River's Edge
Masterpiece
First acknowledgement that i have not done the exact thing you have described. However, i have completed thread grafts on trunks between nodes and then cut back to that point to continue development and change of direction, It works on the trunk. I see no reason why it would not work on a branch, the tendency is to die back to an active growth site be it a branch or node developing. Pruning is usually back to nodes or active branches with stubs left temporarily. Should be a fairly straightforward experiment and with the way that threadgrafts establish fairly quickly, it would not take long to see the results. If you try the experiment, it is a good idea to consider the position of the threadgraft for both direction and location carefully with an eye to design purposes.Got a question, which I think it's not been answered before.
Imagine that I have an Acer with a long node. However, It is a thick branch and I don't want to restart it. Can I thread graft a brach in the middle of this long node and then, when it takes and is strong, cut the remainder of the node above? My question is because usually maple cuts die back until they reach an internode so I don't know if having a grafted branch will halt the die back of not.
Conversely if the answer to your question was no then that would seem to imply that a threadgraft could only succeed if placed on a node. To my knowledge that has never been a consideration for where to place a threadgraft on a maple.