This is a great thread, as is Ryan's sub-topic. Many of us model coal mining and coal hauling, but there isn't much discussion of how we do so.
First, the New River Mining kit is one of the most kit-bashable structures out there as proven by this thread. We bought several of these kits, Glacier Gravel, and the Walthers flood loader kit (Can't remember the name), as well as a few other kits from other manufacturers to get the pieces and parts to kitbash coal loaders on our railroad.
The first picture is of a loader that Ray built, it still needs an extension to the conveyor, some additional weathering, a few more details, as well as ground stuff (Weeds, loaders, trucks, junk, coal piles, etc.), but you can see it is a part of the New River Mining kit.

The second picture is a loader that is still in the construction phase, but again, you can see the New River Mining kit is the source of the major pieces.

Sorry if the pictures aren't of the best quality, I was walking around, hand-holding the camera - some with flash, some just with the layout lighting.
These are two of the largest loaders on our railroad, there will be one larger that will be almost the foot print of the complete New River Mining kit. There are also several smaller loaders of varying sizes located around the railroad. In total there are 8 coal loaders on line.
As for our coal operations, we have followed the patterns we see in the areas we are modeling. We model a freelance prototype called the Roanoke & Western Railway which is patterned after NS/CSXT and their predecessors. The R&W is envisioned to be roughly the same size as the Clinchfield was, in mileage, equipment and traffic volumes.
Since we don't have a warehouse in which to build our railroad (Just half of a basement.), we decided to model a heavy branch with several smaller sub-divisions and extensions that feed coal to the mainline which is represented by hidden storage. We have chosen to model non-unit train style flood loaders since we don't have the length for unit train flood loaders and it is tough, but not impossible, to model both sides of the loads/empties movement on flood loaders.
What we have done is start our operating session with all of the cars at the mines loaded and shoved through or pulled ahead of the loader. When we swap the loads for the empties, we put the empties on the opposite end of the loaders. Since we run live loads of real coal, it's easy to tell the loads and empties when drilling a loader. It also makes the return trip a little more difficult due to the weight of the loads.
Some of our loaders are served by a single train which comes out to swap 15 empties for 15 loads, one example is the Ranger Turn which serves the Ranger Fuels loadout. With some of the smaller loaders we gather up the empties from several loaders and bring them back to the small, 3-track Clay Yard at Black Creek Junction to be combined with other loads from small area loaders to be picked up by a turn from the larger Linwood Yard before heading off the railroad to hidden storage bound for Roanoke.
We always leave enough track on each end of a loader for a loads in/empties out type of operation since this is how the real coal loaders work. We wanted our operators to have the feeling of accomplishing something by spotting those empties at a location on the siding different than where they pulled the loads. The live loads add to this as the train becomes heavier as they make these loads/empties drills, finally needing a pusher to get back up the grade.
We didn't want to force any paired industries just to get that loads/empties cycle, so we didn't model a power plant or other coal consuming operation. We preferred to instead have the loads/empties flow on and off the modeled portion of the railroad. If given enough space, it would have been nice to model either a large coal preparation plant or a coal-fired power plant.
The coal prep plant would have been a big operations booster. The loads from many smaller loaders could have gone there to be cleaned and sized, then shipped back out. This would require yards on both ends for the raw inbound coal and the outbound processed coal on the other end. All this movement of raw and processed coal would require trains to move them around.
To get some feel for coal industry operations you can read two excellent books on the Clinchfield Railroad, the first is Clinchfield Country, by Steve King (Out of print, but you can find copies) and the newer, The Clinchfield Railroad in the Coal Fields, by Robert A. Helm. There are sections in these books that talk about the operations on the Clinchfield and many great pictures of the loaders along the line. Every type of loader you can imagine, from the smallest 1 to 2 car a day, mom and pop loaders to the 100+ a day loaders and the big Moss coal preparation plant.
The other thing you might notice in these books is how much like a model railroad much of the Clinchfield appears. It is often a heavily tree-covered mountainside as a back drop with the tracks right against it, a road, possibly a few houses, and then the othe maountainside on the other side of the narrow valley. Very easily modeled in a narrow space. It was also a very curvy railroad with numerous bridges and tunnels.
Sorry to get off on a tangent about the CRR, but it is a good example of what many of us are trying to model and the books are good references for ides on kitbashing and scratch building loaders.
And finally, a few pictures of our railroad, we only have a small portion scenicked and that isn't finished yet, many details are needed.

Above is a set of R&W units backing down the lead to the Potter Coal Company loader to pull their daily production of 3 loads.

Above is a set of Clinchfield F-Units with loads on the Martins Creek Extension, coming down grade back to Black Creek Junction and Clay Yard.

And this is the same set of R&W power idling away while the Conductor strolls over to the tower to talk to the powers that be.