Thanks Becasse for this exposition. It confirms much of my understanding and adds a lot to it.
When I was planning the model of Dartmouth some 10 years ago an early decision was to have the Totnes & Dartmouth Rly double tracked.
The Directors had some inkling that the tourist trade in S. Devon had potential and wanted to encourage that for their better profit. In reality it was decades before tourism came to Dartmouth and the old trades of heavy engineering, smuggling, piracy and fishing reigned supreme.
Paul Willis had observed that most BG layouts tended to be "pretty". I prefer urban and industrial grot.
The far sightedness of the T&DR Directors did not save the Rly from bankruptcy in 1902, but that is another story, as told on the layout display.
I have chosen to "oversignal" the model for the period for four reasons:
1. My other model of Highbridge in 1913 is under-signalled and my chums give me a hard time over that.
2. The Totnes area was used to trial more modern signalling and embryo interlocking in my period. It seems plausible that many experimental arrangements would be tried on the branch before being attempted on the SDR main line.
3. It gives me an excuse to have lots of different types of signal as Becasse describes above.
4. Its MY model of a railway that should have been but never was. My imagination gets a freer rein here that it could at Highbridge where my attention to historical accuracy is much greater, except for the signalling!
I am delighted that my provocative jingoism was so productive of further insights

The simplified signalling diagram submitted by Becasse was exactly how the ine and station was built in the late 1860s.
My later "modernised" version swept away the old thinking and the line was a pioneer in trying new ideas. Rule 0 applies.