Right: Once the stop hits, its momentum and further furling force from wind both tend to lift the tail off the hinge. Lifting the tail might be how the momentum of the tail's swing is absorbed - in which case if a really strong gust swings it fast enough - or a really strong wind-shift pushes it hard enough once it hits the stop - it will have enough energy to come completely off the pivot pin.
While pinning the tail pivot together solves the probelm of the tail coming off the hinge, it then leaves you with the problem of what to do with the energy from the tail's swing - moreso now that it can't be temporaly dumped into raising the tail on the hinge. This has the potential to bend the tail, both temporarily (spring-wise) and permanetly (yeild-wise, pregressively more with repeated hits). Eventually this could let it bend enough that the blades hit the vane - at speed, since it would happen during furling. (Also: The lifting force is going to hammer somewhat against whatever you use to pin the pivot together. If it's just a cotter pin this might wear it down and sheer it.)
I'd be inclined to add an energy-absorbing (or reflecting) rubber bumper to either the end of that tail-stop or to the point on the yaw bearing where it hits - either in lieu of, or in addition to, any pinning of the tail hinge. That would eat much of the energy of the tail's motion when it hits down, and turn the rest into a bounce-back for another cycle. A suitable bumper for large mills should be available in an automotive suspension or power train: On the front control arms, on top of the tranny, or on the rear axle near the shocks. For small mills you might use something from a car door, or cut away part of the bumper from a suspension.
Adding a projection on the yaw bearing above the point where the stop hits should keep the the tail from riding up the hinge when it stops-out, even if you don't pin the hinge or add a bumper.
