When it describes these scores as “lost”, did the project include the search to locate the scores?
I have for a while been wondering if this model could be used to locate the lost scores of Sergei Bortkiewicz, not least his opera “The Acrobats”, for which there are many clues as to where it might lie - but this might best be done by a professional document hunter, and that costs money. Several thousand euros apparently. I don’t know how much interest could be attracted, certainly his music is stunning.
Many famous scores were thrown away after the recording so the originals exist in a land fill. All of the Miklos Rozsa's MGM scores were tossed. Some of his sketches may have survived (I think they're at Eastman) but not the full orchestrations. Anyone doing those now would have to rebuild them from piano conductor scores or transcriptions. Herrmann's slightly different because he orchestrated everything himself and much (but not all) of that has survived. Virtually none of the Korngold sketches exist; but the orchestrations done for the WB pictures survive because WB saved everything, or nearly everything, and they're at the WB archive that USC administers. These "newly discovered" scores might mean there was a version not known to exist that someone finds in a library or something. All of John Williams' musical sketches will go to Juilliard music library and within them, we might find "newly discovered" works that were just never fully realized, abandoned, first thoughts on what would be famous scores, juvenelia, etc. These days most scores are somewhere digital so can be reproduced easily as long as the disc survives but of course in old days, these only existed in an original copy (unless someone manually copied the score by hand which would rarely be done because there wasn't time and no one thought these would be cherished in the future).