Okay, well whatever the problem was, it was fixed by a second bath in the fix. I had assumed that the a-h dye (or the anti-curl layer) had become partially loosened and had settled at the base of the film in the tank, which is probably technically gibberish. It is clear though, that, for whatever reason, some chemical residue was gathering on the film in droplets and then drying, which when scanned produced the overlapping white shapes visible in the scan. I realised that there might have been a fixing issue when I compared the film to a previously developed film and noticed an obvious difference in opacity/transparency, and both symptoms seem to be cured by the same treatment. I'm not a chemist, so I don't know why or how, but both my specific problems (even if they are unrelated) were solved in this way...