You seem to be internally conflicted about where will is in cognitive science, and so am I. "Will" is not the proper term. From what I understand, the lay term "will" either overlaps or is entirely in the decision-making process; the latter being an actual scientific term.
But I do know what free will is in terms of science, at least the definition I am using for my argument. It is the ability to have chosen differently. It is easy to see that QM effects in the decision-making process could have allowed us to have chosen differently.
Nope.
A random input can cause a different choice for each iteration; but the decision point is in the generation of the random input - it is not a 'choice'.
If I program a robot to navigate a maze using a coin flip - turning left on heads, and right on tails at each junction - then the result at each junction 'could have been' different; but this is not due to a choice. It is due to an external random event used as an input to the choice. The choice itself cannot be different.
No matter how deeply we embed the 'randomness generator' in the robot, it is not able to have chosen differently at any given corner - it can only choose what the random input tells it to choose.
There is no way for neurons to make 'choices' that is significantly different to the way that robot makes 'choices'. I can add other rules - for example, if one path has a red floor, follow that path regardless of the coin flip result - But there is no set of rules, no matter how complex, where the choice can be the result of anything other than pre-determined inputs (even if those inputs are from randomness generators).
You are hung up on "the ability to have chosen
differently"; I am pointing out that free will cannot exist without "the ability to have
chosen differently"; and that there is only the illusion of choice, even if the outcomes of repeated identical tests can be different each time the test is run.