I've done some thinking about the question about H-20.
Before we start, I think it's important to establish that there is no definitive answer as to what the current state or priority of the H-20 project is. For all we know the PLA might surprise us and it'll emerge out of nothingness next week.
However, if we accept the idea that H-20 as we know it has been pushed back, I see a few possible reasons for it
1. There is still a need for H-20 (a large, manned long range strike bomber), however in the immediate term future (next five to ten years) it is not the best use of resources to achieve the long range strike mission.
2. Related to the above, if they are not pursuing H-20 in the immediate future, I do believe there will still be new long range strike aircraft that will be pursued for this decade, which can take on aspects of H-20's mission at lower cost or lesser risk.
3. H-20 itself may have been a highly valuable asset produced in limited numbers. We have all operated under the assumption that H-20 would be a large, four engined stealthy flying wing (powered by four non AB WS-10 class engines) -- such a singular aircraft would be very capable yes, but it would also be expensive and large. Being expensive means that the ability to produce a large number of them will be difficult -- even if they produced 100+ airframes, that ultimately is not that many individual airframes in a large scale high end conflict.
4. H-20 would also likely be somewhat limited deployability/distributive capability. For the PLA, with current geostrategic positioning of US forces in the region, their air bases may be vulnerable to strikes. Furthermore, the fact that such an aircraft would be quite large, means that you will likely be limited in terms of basing and support requirements -- i.e.: it reduces your ability to distribute and deploy your airframes in a more agile way. In other words, the pursuit of a large aircraft that is procured in relatively low numbers (yes, I consider 100+ airframes to be somewhat limited), for the PLA at present may be an unforced error due to the vulnerability of losing them on the ground against opfor strikes, some of which are likely to get in even in context of formidable PLA air and missile defenses.
5. A subsonic, stealthy flying wing will find it difficult to penetrate an integrated air defense in early phases of a conflict due to said IADS being unmolested and un-degraded, which is likely to further contribute to a degree of losses. My view here, is that there is a possibility that H-20s that do get into the air in the early phases of a conflict may suffer some losses if they try to strike mission if the surrounding environment (1IC, 2IC) have yet to be properly sanitized of opfor combat air patrol and if IADS (both ground and naval/ships) have yet to be sufficiently degraded, even if they use stand-off weapons. Of course, it is likely that some missions will still succeed, but the issue is in attrition of said limited fleet size. The answer to this for H-20 may be to simply try and wait it out until enemy CAP and IADS are sufficiently degraded to perform their strike missions, but that leads to the problem of where to reliably base them where they can have minimal risk against opfor strikes.
Summing it all up, my view is that the H-20 concept which we knew, may simply have be a bit too expensive (aka too few in number), a bit too big, and a bit too inflexible (from a deployability pov), and too vulnerable to attrition in the early phases of a conflict, for the PLA to be willing to invest in it at this time.
So, what's the alternative solution the PLA may go for, for the long range strike mission? Well I'm sure everyone will agree that missiles of a variety will be used, launched from ground based TELs, from H-6K/N family aircraft within the relative safety of PRC airspace, and from PLAN ships and future submarines.
However that still leads us to the gap for a fixed wing, penetrating long range strike aircraft which needs to be filled.
My hypothesis (or rather, the solution I personally find most likely), is they will go for an unmanned route -- a large, stealthy, fixed wing UCAV, likely subsonic.
I will call it "GJ-X" for now.
GJ-X would likely be somewhat larger than the GJ-11 -- something that's bigger than X-47B or S-70, but powered perhaps by a single uprated non-AB WS-10 engine (as opposed to four such engines on H-20). Such an aircraft would aim to have an unrefuelled combat radius of being able to launch a strike mission well over the 2IC from operating within central China, and able to operate autonomously in certain phases of the mission but the CONOPS would also involve semi man-in-the-loop for parts of the mission by either a J-20A or in future the J-XD next generation fighter (which is likely to have even greater range than J-20/A), that would operate many dozens or hundreds of kilometers behind them to control specific parts of the mission.
In terms of the size of GJ-X, it would be far smaller than a H-20 and H-6, and perhaps somewhat bigger than your average Flanker (shorter but greater wingspan).
In terms of payload, this "GJ-X" would aim to be able to carry perhaps 4-6 tons internally (less if it is conducting a longer range mission), with primary payloads being high performance stand off A2G weapons, but also a full range of guided unpowered weapons to conduct direct attack missions against a degraded IADS. (other payloads like AAMs, smaller UAVs etc may be secondary/tertiary payloads and I mention it for the sake of completeness). It would be equipped with appropriate sensors and datalinks for the strike mission as well.
Now, GJ-X would obviously carry less weapons than a single H-20 would, however four GJ-Xs may carry a similar amount of weapons to a single H-20 (four GJ-Xs using four engines, while a single H-20 uses four engines as well), and a GJ-X would have a similar level of stealth to what H-20 would have aimed for.
A single GJ-X would of course still be a somewhat expensive aircraft, and it would not be intended to be "attritible" but it would be an aircraft where losing one or two dozen of them is not going to cripple the availability and sortie rates of your overall fleet. Being an unmanned aircraft, you also may be able to fly it less than a manned aircraft (less need to train pilots regularly), and other life support costs and needs on the aircraft itself can also be cut due to no pilots, all of which can help to offset the costs of producing more airframes.
For me, I envision 400-500 GJ-Xs being procured, instead of say 100+ H-20s.
Importantly, the smaller size of GJ-X would allow far superior dispersion and deployability capability than something like H-20 in a manner (it may be not worse than a large stealthy tactical fighter like J-20), while possessing the range to reach out to 2IC, and possessing the numbers to shrug off early phases of attrition when they conduct strike missions, and still having sufficient size for a single aircraft to carry 2-4 large standoff payloads internally. GJ-Xs may also be sufficiently small to enable semi-hardened or even hardened shelters to be produced for them, to further mitigate their vulnerability (such shelters would likely be impractical for an aircraft of H-20's size)
With in flight refuelling, GJ-X may also be able to reach out beyond the 2IC to Hawaii (depending on where it is refuelled, and depending on the earlier phases of the conflict -- obviously it's not something they'd do at the opening rounds).
However there are some caveats:
1. GJ-X would of course be unable to have the absolute range that H-20 would have on a single aircraft -- GJ-X would be optimized for targets at 2IC distances (unrefuelled) or Hawaii (refuelled). However, considering where a conflict is likely to be fought for the PLA, focusing on the western pacific is probably the higher priority right now.
2. GJ-X would be unable to carry a single large payload if H-20 were able to do so by virtue of being a smaller aircraft. Of course, it's unclear whether H-20 would have been able to do so either depending on what its weapons bay geometry was, but I digress.
3. As a UCAV, GJ-X would of course be vulnerable to EW, datalink interference and so on, therefore in theory this would be a risk to its conops, however I think given the way everyone is pursuing higher end UCAVs for all elements of future air power, this is something that everyone will just have to accept.
4. GJ-X is far from the only type of high end UCAV the PLA will pursue in my mind -- I can see them procuring a large number of (smaller) GJ-11 sized UCAVs focused on 1IC to 1.5IC distance missions, as well as other flying wing UAVs for the ISR mission (potentially variants of GJ-X and GJ-11). And this obviously doesn't include A2A MUMT UCAVs of which we can expect multiple types to be procured.
In short, what I am envisioning as a likely alternative to H-20 for the "fixed wing penetrating, long range strike aircraft" role can be described as:
"Aerial distributed lethality, optimized for 2IC to central pacific distances"
