Preemptive strikes are an option that would come under the overall banner of offensive counter air.
The number of aircraft that the ROCAF can sortie before the PLA strikes their air bases will depend on how quickly and how prepared each side may be when hostilities are joined, and the exact political circumstances of how the political circumstances evolve to a declaration of war can be really quite varied and not too relevant to the nature of OCA itself.
Additionally, any ROCAF fighters that do manage to sortie initially will likely have their air bases either newly attacked or subsequently re-attacked to maintain them in a mission-kill status.
So for the factor of "sortie rate," some of the important questions are how many aircraft can get in the air initially, and how many aircraft can repeatedly land and get back into the air in an effective and mission capable manner.
In a way what you described is superfluous to the V upgrade because those overall considerations are independent to whether Taiwan did or did not undergo the F-16 upgrade. That said, the improved capability does bring to Taiwan additional deterrent capability and impose more complex calculus onto the Chinese planners. It will more likely than not allow Taiwan to be in the game longer and provide more time for the US to mobilise and to respond. Without air domination, China is unlikely to risk invasion.
The kill chain itself as described and portrayed by the original paper emphasized the use of AEW&C to provide an initial firing solution and mid course guidance for the missile by datalink. Considering the point of an AEW&C is to actively emit, using an AEW&C to provide the initial guidance for a long range AAM is not much more different to having an AEW&C actively emitting in a battlespace in generall.
By the terminal phase the missile will have its own terminal seeker (RF and ImIR combination supposedly).
There are two things that I don't really get it based on your description on the use of AEW&C in providing targeting solution.
(1)They would probably be subject to similar risk as their intended target unless they are placed well behind area of operation. Have you done the maths that this could actually work - like detection range and range resolution?
(2)Data linking requires robust network with nil latency. Is China there yet? I don't think the USAF even has this type of CEC capability. The USN has made further progress than the rest of the other services. A link 16 type of datalink will not work for cooperative guidance because of latency.
Of course the paper was written quite a few years ago -- I wouldn't be surprised if in the future the integration of being able to hand off datalink to more "forward placed" tactical fighters or stealth fighters would be on the cards -- i.e.: not too different to how the US has looked at using F-22s and F-35s to guide in AMRAAMs launched by non-stealthy fighters or to guide in SAMs launched from ships like SM-6.
Within the F-35 community, cooperative engagement could be executed because of MADL. Between the F-18, Aegis and E2D it is because of NIFC-CA and the TTNT waveform. It requires big data pipes and building those is heavy investment. It is not by way of Link 16.