This is a very valid observation and one I think might be applied to a few topics of Fortean nature.
I think, in the next few years, rather than turning up new material, the great advances will come from meta-analysis of existing work.
The ability to digitise material, gather from disparate sources and analyse not just sources but also methodologies, will allow greater insights than ever before. Advances in machine learning and AI should allow greater application of this kind of work, not to mention expediting it.
The JFK case, and even the likes of the Loch Ness monster, should lend themselves to this kind of study, where vast amounts of historical material can be analysed for value and veracity before being included, or not, in the canon, but weighted accordingly.
All we need are a cadre of PhD students with Fortean interests, access to high performance computers (HPC) and susceptibility to bribery through snacks.
How hard could it be?