To cross the informations you can get here and there with your personal observations is the way, the ONLY way to go.
I am trying to do the same job with the GMTs, and I find it easier to do with the 6542 than with all the variationsof the 1675.
As the 6542 had a production of only 5 years, and not many were made.
A safe approach is to stay stuck in these production years, for every Viintage Rolex Watch, if you don't want to take the risk to fall on a re assembled watch.
The golden rule is to consider that the more you are on an exotic watch, the more you take the risk to get a re assembled watch.
That being said, yes, I agree with you on your method, classification and datation.
With a small difference when it comes to 1675... The serial number doesn't lead you to the same datation, sometimes, when you compare a 5512 / 13 with a 1675.
For example, my year of predilection, 1964, comes with a slightly lower serial number, for the 1675. As I saw some with a 9xx xxx number.
And personaly, I would not worry to see a Gilt 1675 from 1967, while I woulmd worry to see a 5513 Gilt from the same year.
Another difference is about the underline dials of the GMT and the Sub.
On the GMT, you have basically 3 years for underline dials: 1960 / 61 ( most often 61 than 50 ) and 63.
While it seems that underline dials didsn't come as early on the Sub.
So many things to say, to observe and ti learn.
Thanks for bringing your stone to the edifice, Bill!