Swamp,
This is a question that I've given a lot of thought to, mostly because I waste so much time and effort getting it right! I believe that the shape of the taper on the bow tip was traditional, going back probably thousands of years. People back in the middle ages were practically-minded, but also very tied to tradition - if something worked they were unlikely to change it. With respect, the straight taper works because of good modern glues, whereas the gradual taper works even without the horn! When I do this, I rough out the bow-tip to what looks a close match to the profile of the drill bit, then I drill out the horn and use this, placed on the bow tip and rotated round and round, as a way to find the bits that need further reduction - the horn leaves a mark on the wood. I have also wondered whether the horn nocks might have been heated up to soften them, before placing them on the bow tips (with or without glue) so that when they contract and harden up again they would form a tight fit. As you have suggested, they might have used a stone grinder with the exact internal profile of the horn, to work the wood down. It would be interesting to see if the cones on the Mary Rose bows were standardised in any way from each particular bowyer; this might indicate such a system. However it was done, it was likely to have been done using a quick and efficient system to ensure a rapid production rate. Certainly this idea doesn't tally with the way I personally do the bow nocks, but then I am probably unusually slow and incompetent here
Cheers, Dave