I would say it depends on the size of, and amount of commonality between, the two datasets. If you have two registration forms, each containing dummy values in hidden fields for the data not applicable to each registration, although it may be a little less efficient in terms of data storage, you should end up with just one relatively simple (if big) function that handles both registration types. If you want to post an overview of the datasets involved, I'll try to offer a more definative opinion.