Also, is it necessarily fuel costs behind the decision to steer away from 747, or could it be other costs, like plane utility and capacity, parts and service cost, and the requirements of cross training?
If an aircraft even has the capability to hover in place, it usually has to expend even more fuel to do so than flying an orbit or holding pattern. The ideal way for an aircraft to handle the counter-gravity force is to push an infinite amount of air downward at infinitesimal speed. A helicopter would want to have infinitely long rotor blades, rotating at an imperceptibly slow rate. Obviously, we can't do that, so very long blades spinning slowly enough to see them move is the best compromise we can manage. The output of a jet engine is a relatively small amount of air, moving very fast, but the jets push the craft though the atmosphere, and the wings translate that forward motion into pushing a larger amount of air downward just a little bit, for more efficient lift. Vectoring the engine thrust downward to hover in place means you lose that efficiency boost from the wings. Also, rotary-wing aircraft that are heavily-laden experience "coning", which reduces hover efficiency by making the force from each wing slightly off vertical, and some of the lifting energy goes into canceling out a lateral wobble.
Keeping a plane in the air for another 15 minutes of fighting gravity just isn't significant compared to the effects from changing its speed away from the aircraft's efficiency optimum in order to arrive 15 minutes later, or from fighting air resistance by doing a holding pattern at optimal speed and altitude for enough extra distance to expend another 15 minutes. As long as the flaps are not extended, the plane is just not very concerned with gravity.
Also, once you hit about 11 km/s, you don't need to expend any more energy on counteracting gravity, anyway. At that point, it's all about atmospheric drag.
An approximate description of a train.
For many vehicles, the difference in that factor between running at ideal cruising speed (the 60MPH car in both of our examples) and a lower speed is minimal, and could well be offset or overtaken by the extra spend in (2). This is my 30MPH car example.
Running at a higher than ideal speed is inefficient in this factor and increases fuel usage a lot. This is your 120MPH (my 90MPH) example.
As you say, airlines care deeply about costs. It is highly unlikely that they were mostly running their planes inefficiently above design cruise speed, and are now slowing down to at or below the design cruise speed. It is more likely that they have always been at around the optimal speed, and are now going slower than that to reduce the risk of fines.
Fuel expenditure falls in to two buckets:
1. Moving the vehicle. Accelerating to cruising speed and fighting the deceleration of friction and air resistance.
2. Keeping the vehicle running. Lights, climate control, that sort of thing.
My guess would be that the vast majority of fuel usage for planes is (1) not (2). As evidence of this, airlines care deeply about costs. It's not a high margin business (any more). My argument is that if it wasn't overall more fuel efficient to fly slower airlines just wouldn't do it. More evidence, the latest planes (A350, B787) have a lower cruising speed than, say, the 747. This also wouldn't be the case (IMHO) if it didn't decrease fuel costs.