Potential problem: The "Remote Type" for the Zip Codes is different. One is a floating number, the other is a character string.
Problem Solved: I have to change the data type consistently into a string or a number despite them both being recognized as zip codes.
One problem to still be solved is that I want more of an "outer join" if possible. I can only use all the zips that are in the primary dataset. I want to use all zips in the primary and secondary and overlap. But I can live with this solution for now.