But you're right that I can delete the fields I don't want, so maybe that's a way to go.
That's not the real issue. Your library already contains lots of data you're probably unaware of. It either is essential for the proper function of the program, or has no significant impact on performance. You'll achieve the greatest efficiency by ignoring such things.
The real issue is the data cannot be directed to fields of the user's choosing, or omitted so they don't overwrite data from other sources. These capabilities are essential if the standard fields otherwise used are inappropriate to the user's needs, or one wishes to combine data from different sources.
But these capabilities are beyond the scope of a simple "lookup" function. What's needed is a meta data management system that allows data from a variety of sources to be mapped to any fields according to overwrite preferences. Such a system could also provide for user-specified expressions for each field for reformatting, revising or filtering the original data.
A meta data management system is needed for a number of other compelling reasons: So different sources can be specified (using rules) and configured for different media. So meta data can be obtained automatically when files are imported. So certain media types can be updated according to a schedule...
Hopefully, we'll see something like this soon enough that there would be little point in adding functionality to the existing lookup tools.