I didn't misunderstand anything.You mis understood my statement or I wasn't clear.And on parallel trial. Coma line is well suited as much is flat on or near natural ground.In terms of gradients for cycling, all adhesion railways are flat. The steepest railway ascents in Australia would never see any colour other than green (0-3%) if you prepared a TDF-style gradient profile, and even the legendary routes in Switzerland and Austria would only see occasional stretches of the blue (3-6%) for a couple of kilometres at a time.
That's why rail alignments make such great cycling routes for commuting (e.g. the Coast to Vines Bikeway in SA, a far greater success than the railway to Willunga that it replaced) and touring (the market for a Queanbeyan-Bombala route), but they are not so great for sport cycling where the interest is in seeking out the tough climbs.
If you want to operate a rail trial and a working railway, even an irregular used one like HR, you need to have a path about 2-3m wide alongside or near by the railway ROW but with a safe buffer. Which means the combined ROW is quite large and basically a twin track with one line missing like the cycleway built in Hobart. However this line to Cooma is single track, but the surrounding terrain is also relatively flat compared to many other lines where the railway is an embankment or in a cutting of which the older cuttings are rarely wide enough for a bikeway to built along side complying to modern safety requirements. Thus for the majority of the parallel route if the existing Cooma Heritage Railway operation, it wouldn't be a huge cost to build an adjacent trail.
I appreciate the rest of the detail provided though.
My point is that, in general, the occasional small dip or hump that results from following the terrain away from a cutting/embankment is not significant for cyclists.
Keep the height difference within about 4-5 metres and keep the gradients to a medium level (5% max) suitable for touring cyclists and it's not a problem.