You raise a good point, and perhaps I need to be less serious about this - modelers are entitled to use whatever term they like. The engineer in me wants to be accurate, perhaps too accurate, but then the rolling characteristics of roller bearing versus "plain journal" bearings was quite different, and that has operational implications for those trying to accurately model STMFC operations.
I used to be impressed with the idea of a steam locomotive on a model railroad crawling away with a freight in tow at 1 smph. Big flywheels, great decoder, etc. But I would now challenge anyone to show me a movie of a steam locomotive sustaining a 1 mph speed for any significant distance when starting a train. Even though slack action was frowned upon, when starting a freight train with plain journal bearings, the engineer really had to take slack, reverse, and then quickly get the locomotive up to at least 2 to 3 mph until all of the slack was gone and the entire train was moving. They would have to increase the steam flow to accomplish that as the train started, but you didn't want to be any faster than 3 mph until the caboose was moving. Then and only then would the train accelerate.
Personally I think that would be a neat thing to model but YMMV....
With all roller bearings in the consists today, the need "take-slack" and then to "walk-out" the slack disappeared, so I suspect that part of the art of being a steam locomotive engineer has been lost.
As for layout owner terminology, to each his own I guess, but then I know very few PRR modelers who call their cabins "cabooses" ;-)