I am confused as well. I asked a question at reddit and a kind guy attempted to answer (it was before the 7.3 "aftershock").
This 2012 study investigated the coupling in the area. You can see that coupling is complete at shallow depths throughout the whole zone. You can also see that coupling tends to decrease at depths towards the eastern termination of the rupture zone of the mainshock, and that coupling was increased in the area of maximum rupture during the mainshock (near Kathmandu). Further towards Bhutan coupling increases but variations are smaller than in subduction zones.
This paper investigates ancient earthquake scarps in Bhutan and tells us what a truly big earthquake looks like in the Himalayan region.
If you see the USGS page, aftershock locations are seen to expand outside the rupture area. There are 2 strong aftershocks of particular interest, both around 6.7 magnitude: one at the north-east corner and one at the north-west corner. The fact that they are towards the north is worrying since this is where coupling changes rapidly from nearly 0% to nearly 100%. As Bilham said, during really strong earthquakes even the deeper uncoupled patches slip and add to magnitude. The western aftershock was in an area of minimal slip during the initial earthquake and this surprised me despite the fact that towards the west there were less strong earthquakes in the recent past (so it theory there was room for a strong one), while towards the east there was the 1934 earthquake (which however was not that huge, and additionally in the 80 years since then a couple meters of convergence have taken place, a significant amount of energy).
If you see the 1st link (figure 7) the most suspicious area is a narrow zone near the location of the 7.3 "aftershock" (and near the eastern aftershock as well). However the second most suspicious zone is in western Nepal. So, there are arguments for both sides, the one claiming that towards the west there is room for a large earthquake, and the side claiming that towards the east there is a large possibility for a perhaps less severe earthquake. The eastwards rupture (that compressed waves and perhaps stimulated more fault patches than in the opposite direction) and the uneven distribution of aftershocks favored the eastern side from the very beginning.
The history of Himalayan earthquakes reminds us that 10% of strong earthquakes were preceded by foreshocks or were stronger-than-usual aftershocks (like in this case). However, to my knowledge, the stronger foreshock even recorded was 8.1 (or was it 8.6) in the case of the most powerful earthquake ever recorded. A Mw 7.8 being a foreshock of something stronger would be very unusual, yet from the first day experts were warning for potential giant shocks near Bangladesh. This was in my opinion not correct. Sure the lack of a truly large earthquake in the whole region is unresting, but why now specifically and why there?
Unfortunately I cannot add more links due to being new here. I hope that, even if not helpful, at least I assured you that you are not the only one in confusion.