That's about what I go for as well.
But I'd go with a 100 J-10C delivered over the next 8 years at minimal production levels at Guizhou, rather than in just 3-4 years.
And I'd skip the J-10A upgrade as they already just went through their MLU.
The J-10B are about due a MLU anyway, but they're already similar enough to the J-10C, so I'm not sure what you would want to upgrade?
I'd also keep the Su-35 around, since they are new and also capable.
PLA has gone through this period where they had a lot of small incremental improvement types to finally at a point where their product is as good or at least in the same sphere as the best in the world. Think J-20, Type 055, 052D, Z-20, ZTZ-99, Y-20 and 054A. Then, they can really start mass production. During this time, they had to go through a lot of incremental blocks which adds to cost. Now that they have arrived at a comfortable point, they can work on lowering costs by trying to reduce their fleet complexity.
As such, things like Su-35, Su-30s and J-11As should get removed from service as soon as they can be adequately replaced. Su-35s still have a lot of time left by 2035, so they might just want to sell that back to the Russians or someone else. J-11Bs that get upgraded should have similar avionics as J-16 to reduce the complexity of maintaining that. J-10As that get upgraded should have similar avionics as J-10C and use WS-10B. Similarly, J-10B/C that still use AL-31FN should be modified to use WS-10B. Fewer subtypes means supply chain will have to maintain fewer product lines. There are more spares around. All good things. All factors that will lower maintenance and improve availability.
We keep talking about numbers. numbers and numbers. Do you know what helps numbers? Availability and lower maintenance cost. You get the same air time with 500 J-10s available 70% of time as 700 J-10s available 50% of time. Why did IAF have such horrible availability with Su-30? It was a foreign product that was not in service with the exporter. Supplier chain was further away and not providing enough support. There were enough spares around. There were a lot of different subtypes in their Su-30 fleet. It's in PLAAF's interest to reduce small sub fleet, subsystem differences, foreign reliance of engines. Getting rid of Su-35/30/J-11A entirely would allow them to not have to support Russian missiles anymore. Similar, you want to replace the original J-20s with WS-10C engine at some point.
A lot of the J-10As are still quite young (some are under 10 years old). There is no rush to upgrade them. Get Guizhou working on producing new J-10Cs for PLAAF right now. Once that's over in 3 or 4 years, they will have some export orders and a lot of upgrades (think 200 J-10A/B) to be busy until well into 2030s. Even the early J-10Cs would need an engine switch at some point. If the goal is to keep Guizhou working, then my proposal will keep them working. If the goal is to keep supply chain well stocked, my proposal should also do that.
By end of 2040, you could be looking at a number of 6th gen aircraft, 1000 J-20s (400 with WS-10 and 600 with WS-15), 400 land version of J-35, 350 J-10s of various types and 350 J-16s. That would be very modern and much reduced in complexity. It will also be a larger fleet than now.