This is bizarre. I literally just told you that this is a misrepresentation of my views, yet you persist in it. So that puts me into something of a dilemma: either you genuinely don’t understand what I’ve been trying to tell you, or not. In the first case, I’m running out of ways to explain this to you; in the second, any further discussion would obviously be pointless.
But I’ll give you the benefit of the doubt and assume that you’re still just genuinely confused on the issue. So let’s make this simple. Suppose I’m hungry. That’s about as simple a biological want as there could be, right? Suppose there’s an apple on the table before me. I grab it, and have a bite. Obviously I do so because of my hunger—because of my biological want, because of my genetics, what have you. But does it follow that I do so solely because of these factors? No, not at all!
Suppose then that before me, there is an apple and a piece of chocolate. I have a strong preference for chocolate, which I’m sure is due to a combination of genetics, socialization, and whatnot (it’s a high-calorie food, I’ve associated with good times in the past, etc.). I take a bite out of the chocolate: obviously, those things are a part of the reasons for this decision!
Wind back the clock. I also have a preference for not getting any fatter, also for reasons of genetics and socialization and the like. So I spurn the chocolate and go for the apple instead. Again, plenty of reasons for that!
Now, your view is that on balance, in every given case, the total sum of these preferences fully determine the choice at hand (or maybe, if there is a residual uncertainty, some quantum event might cause a neuron to fire that tips the balance one way or the other). And this might be the case!
But it also might not be. There might be a genuine open possibility here. Then, rather than quantum randomness stepping in, I simply propose that it is the agent’s free choice that adjudicates between these options; that it is the further reasons that single out one option over the other that are due to the self-determination process of the agent, however that might work.
So there is no problem for this notion of free will, at all. The idea that somehow I’d require that all reasons for acting are due to my own self-selection, that my hunger has nothing to do with me eating, is simply absurd, but also, that there are such wants doesn’t pose a problem for my model at all.