He is unable to accept for example, without evidence or at least a plausible pathway, that the superior oblique muscle and trochanter of the human eye could evolve through natural processes.
Is he aware that, when the basic arrangement of the bones and muscles was first developed, it was in creatures whose eyes pointed out to the sides (and possibly also upward) rather than forward, and whose "frontal" bones (which the trochlea attaches to) were roughly horizontal and located between and behind the eyes rather than above? In
this image, to see where things were back then, you have to picture the eye rotated "clockwise", the muscle's insertion point on the eye swung around "left" and "up" (more medial and anterior) with that rotation, and the trochlea "lower" (more posterior). That takes out the U-turn.
Another way to look at it is in
this one, in which you have to picture the eye rotating on an axis that's vertical within the picture so the side nearest you (including points 5 and 11 and the superior rectus's insertion point) goes "left" (back) and the other side (the one you can't see) goes "right" (forward). That would also move the superior oblique's insertion point farther from you as the viewer of the image (medially) and then "right" (forward), passing "behind" (medial to) the superior rectus's insertion point, while the frontal bone leans over to the "left" (back), taking the trochlea "down and left" (down and back). It's another angle from which to see the same thing: the U-turn is naturally taken out by the changes in the shape of the head as you visualize starting with a human head and making it more like our ancestors'.
Of course, this is the reverse of the actual flow of time. What it really means is that if you start with a flat-headed ancestor that has outward-pointing eyes and no U-turns a few hundred million years ago, then crank its eyes around to point forward and shove the top of its head up to make a forehead, the superior oblique's insertion point on the eye moves back and out (laterally) while the trochlea holds on to the middle of it on its way up and forward, thus creating a U-turn in what had been a perfectly ordinary linear muscle before.
I think the annulus of Zinn would also have been farther forward back then than it is now, which, if that is the case, would also contribute to the same effect as it migrated back.
It's pretty much the same deal as the biceps brachii being wrapped around from one side of the limb to another because the original orientation of our arms was like pointing our elbows out, back, and up, with the palm inferior and the thumb medial... or the sternocleidomastoid crossing over from front of the chest/neck to the back of the neck/head because both ends were closer together back before we tilted our heads roughly 90° off from the original orientation.
I just went through all of that because, in my experience, people who ask about an individual structure like that are usually trying to picture how it arose from nothing in an otherwise identical critter, rather than thinking in terms how that structure fits in the big picture and how the whole context around it was changing at the same time. Of course, if he's already thinking in terms of how the whole head and the orientation of the eyes changed together, then he already knows the muscle was straighter originally, so I don't know what the mystery is. Is it just why one of the muscles originating at the annulus of Zinn would have been snagged by a piece of cartilage while the others weren't?
Maybe back when the muscle was still mostly in-line but the changes around it were just beginning, a slight adjustment of that muscle's angle of action was selected for while the difference was still incremental. Or maybe they all went through the trochlea originally and the others came out of it but that one didn't make it. Or maybe the trochlea was left over from having served some other purpose originally. But that seems to be a minor trifle to me; I'm thinking the U-turn shape is the issue here, and that's not hard to see coming from an original straighter shape as I described above. In fact, given what we know about the changes in the skull since our earliest tetrapod ancestors, the fact that the superior oblique was once straighter is pretty much inescapable anyway.