What an interesting post! My approach to cellulose for years has been to cut it back with 1600 or 2400 grit paper a week or two after spraying and then polish up using T-Cut, before finally applying Turtle Wax. I usually find that after perhaps 6 months or a year a slight "orange peel" effect shows so I repeat the process. Subsequently, it is Turtle Wax alone, usually applied after each time I wash the bike. Some parts treated this way have lasted for 25 years and still come up looking absolutely great. Other parts have needed to be resprayed from time to time. My experience seems to be that if I can get the paintwork looking good then leave it alone (apart from some Turtle Wax), all seems to be well. But the T-Cut definitely softens cellulose paint although it does seem to dry out later. I have often thought of changing from Turtle Wax but have stuck with it because of the good results I have had. I find that modern fuels will attack cellulose around the filler cap but that their effect lessens in time and the tank panel on my RH500 (painted in Nov 1990) seems completely impervious to them.
On the other hand, Madam's 1927 Standard is painted with I don't know what. Washed down and with a coat of Turtle Wax it looks good, but there develops something like partially spherical bumps maybe 2mm diameter which are only visible on close examination. Washed down and retreated with Turtle Wax next time round they disappear for a while, then come back. I have put this down to me using a different wax from the previous owner (although rumours abound that he was, shall we say, economical with his use of any polish).
So using the wrong things with my spraying seems to have done me little harm, but it has on Madam's car. Please don't mention this to her!
36 4G, 37 VH, 53 ex ISDT KHA (project), 54 KH(A), 54 4G Mk IV (project), Healey 1000/4 (project)
Former Machine Registrar & Archivist, General Secretary and Single Spares Organiser.