I think there are some complexities here. Please correct me if I am wrong but I am writing from memory and some conjecture from my side so happy to be corrected.
The J35/31 project till recently has been a private venture. That means the Chinese had not committed any resources to the project. To be fair , even though there are rumors of Chinese Government involvement, no one can say with any certainty whether this has happened. The baseline model had already been developed so the assumption was to help the Chinese party develop their product without any sovereign Chinese funds being made available. Now this is where the problems arise. Firstly it is a project without any known engine based on a very inadequate RD93, with no suitable 5th generation engine in sight. Even now PAF has yet to officially commit to a Chinese engine and news keep coming out of J10 being acquired with AL31 engine (I am merely writing this as no confirmation has come through as to what powers the Pakistani J10s). We do not know what systems have been developed and the Chinese could well turn around and deny us the AESA radar/ other hardware of the J20 on the basis of the project being a private one. Other providers will be reluctant to provide help for fear of leakage of information to the Chinese
PAF has felt all along that it will commit to a project once it has seen a commitment from the Chinese Government as that opens up all avenues for the platform to partake of any hardware that has been in use by the Chinese. Normally PAF will try out a Western hardware and ask the Chinese to bring out a comparable one for the JFT. Where one is not available they have gone for a hardware from other providers till the Chinese provide you with one of their own. 5th generation technology is complex and I am unsure EU products would be wanted for preservation of the secrecy of the platform. The Turks will be out as it is none of their business what PAF does as long as they are getting nothing out of it. So PAF avenues for help would revert back to the Chinese giving them the power to dictate what they provide at whatever price they want.
With a prototype already developed and perhaps not to PAF's liking they might require changes which would then open a Pandora's box of time, money , manpower, repeated testing. The cost escalations in that case would be something which PAF would not be able to bear.
Instead PAF has kept itself open to suggestions, liaising with the Turks and the Chinese as and when required. Various subparts/Hardware can be acquired from wherever it becomes available and PAF still retains the intellectual rights over the project. In case it encounters difficulties it can negotiate with either the Turks or the Chinese to get on one of their projects which hopefully would be developing by then. The cost would be less and in turn for giving up local research the PAF can demand a higher chare in the development of the platform with some workshare agreement. In view of getting rid of competition it would seem like a sustainable sacrifice for any provider.
So in short I think the decision is a good one and will bear fruits in the long run. I don't see production starting till the mid 30s but the overall outcome would be better for PAF than not having anything in their hands.
My 2 paisa's worth.
A