Here are some pictures. One thing to clarify is that I do not use the same method of closed loop gravity/siphon transfer that the OP uses. I tried it a few times and it would always stop half way through. I think there is too much restrictions in my lines and disconnects that slows the transfer down and stops.
The same thing happens to me sometimes.
I think what is happening is that there is not a sufficient pressure differential between the 2 tanks to overcome the restriction from the 2 liquid dip tubes plus the 2 QDs plus the liquid tubing once the level in the FV starts to drop. In other words, the pressure exerted by the siphon is no longer enough once the higher tank level drops, and the lower tank level rises.
I've tested 2 different solutions, and both worked about the same:
1. If you are have enough pressure stored up (say >5 psi), burp the SK's PRV to give it a lower pressure relative to the FV.
2. Add external CO2 pressure to the FV. For me it only took a few seconds of gas to kickstart it.
If you could create a larger vertical displacement you could also help yourself.
There may be something else going on here because the gas-gas jumper should in theory equalize the head pressure, but I've clearly had situations where it wasn't.