Is a lot like octane improvers, add to a tank of gas and it will improve 87 to 87.7. 89 is cheaper. OTOH if marginal on 93, or using a turbo in the mountains then it probably won't hurt.
Drag goes up by the square of the speed. Figure I remember was that it takes 15 hp to move a 73 Impala down the highway at 60 mph.
By far, the major drag is from pushing the front of the car through the air. There are three major components of drag, the density of the air (which you can't do much about though a forward facing plasma cannon can help, anyone notice XXX (XXX- state of the union) firing into the water just before he hit ? Was to disrupt the fluid density), the frontal area of the vehicle, and the drag coefficient (cd) of the front.
CD is important. If a barn doow is a 1, most SUVs are in the .3-.4 range and a very slippery econobug will be under .3 (Prius is .26 - is a reason for the odd shape. My heep is about .37 with a cdA of 11.6 sq. ft. See
here for a fairly old list.
Now most upright travel trailers have a cd in the .8-.9 range and can double the frontal area of the TV. OTOH my TrailManor is almost entiely inside the slipstream of the heep. Roofline is almost the same (one reason I do not want to raise it) and just a couple of inches wider.
That said there are three places where turbulence occurs:
1) Under the Jeep (and an air dam & skirts could help)
2) Between the TM and the Jeep (could fill with panels like you see on big trucks. Recently I am even seeing underbody skirts on some
3) At the back of the TM (where some ducting surfaces might help)
However all three are minor for us compared to the drag of the nose of the TV and if you really want to increase MPG without simply slowing down, that is where to start. Can't go overboard though because that is how cooling air gets to the radiator.
So short answer: can "vortex generators" help ? Probably but you may not be able to measure the difference without a smart phone and a timed course.