Many of that is absolutely true.
But one other variable is how large the holes are drilled in comparison to the size of the sholder of the bolts. (Also how square the holes are to the plate). Also if you have the sholder of the bolt through the three plates and not on the threads which break down and make slop which give room for the plates sliding.
If the holes are very tight to the bolt, and the shoulder sits through all three plates, it is very similar to the press fit dowells that you are speaking of. Having the holes sloppy to the bolts, and the threads in the area of the plates does allow for the movement you are discussing.
Art is a precision guy. I can bet that his set-up is basically a light press fit set of bolts in that plate. But just looking at it and then copying it with sloppy holes and improper fit would not be a ton stronger than stock.
Again sounds good in theory, but not really true.. If that plate bent .015 forward it might only distort .002-.003 thousandths in a linear direction in the area that the bolts are in. (Which I can assure you he's more then that for clearance for bolts.) The bolts will require some clearance because if they were line - line you'd never be able to get them in the holes... That's why they need to be a "press fit" there can't be any clearances if you want a truly rigid setup. I.E. Pressed in 1/2 inch dowels..
At the end of the day it doesn't really matter.. If the objective is to get the setup to be rigid and not to move (at all) then adding some press fit dowels around the plates will greatly increase the odds of zero flex in his setup. If the objective is to have the plates not move beyond this Point __________ and his current setup does that... Well then he's succeeded in his task.
Like I said I dunno shit about the setup of a V-Drive nor do I pretend to know. I just know from machining over the years what's "rigid" and what is not, and currently that's not much more rigid then just the single plate. Just beyond a certain point it won't bend as far as a single plate but it will bend with minimal increase of force to a degree.
RD