I guess I would be well advised to fully study the examples provided before responding to the question, but instead I have decided that an immediate answer might have its merits too ... so here goes.
I think that there is more than one mechanism which can initiate a turn but that by far the main one is countersteering.
This causes gyroscopic precession of the front wheel with the result that it wants to tip into the turn by rotating about a horizontal axis at 90deg to the wheel spindle.
However, the front wheel cannot rotate in this manner because it is constrained by the tyre's contact with the ground and therefore it rotates about the contact patch instead.
Other inputs to the bike, like weight shifting or weighting one peg (essentially the same outcome via different methods) also contribute to the initiation of a turn, but these are supplementary to the main, countersteering input.
I guess what I'm saying is that although forces may want to act about their centres of rotation (spindle) or of mass (C of G) etc, the resultant outcome of these forces is that rotation must occur about the tyre's contact patch, because that is the only available option (as long as the wheels are on the ground).
Right ...... now I'll study the Tony Foale stuff and see where I end up after that.
I do very much respect the man and I reckon it will probably turn out that we're actually saying much the same thing in different words .. or else any error will be mine.
But for the sake of the discussion, I've decided to shoot from the hip in the first instance (and did likewise in my original post too) ... I'm hoping that it will make for a better learning process all round, myself included.
ps. actually, gyroscopic precession forces wont be about a horizontal axis but one which is mutually at 90deg to both the wheel spindle and the steering axis .. but it would have got too wordy to include that bit in the above text)
Dare I press "submit reply" now, I ask myself ?
Yeah, bugger it.