The normal reason that a CD isn’t found by the Brennan when there is a entry in the Musicbrainz database is that the entry in the database does not have a disc id associated with it, unfortunately this is very common.
When the Brennan rips a CD it calculates a disc id based an the total length, number of tracks and individual track lengths. It then uses this disc id to search the Musicbrainz database. If the entry in the database doesn’t have this disc id associated with it then the lookup will fail.
This method requires that the CD you have has to exactly match the entry in the database. Musicbrainz does support an additional lookup mechanism (fuzzy search) where it can find entries in the database where the track lengths closely match which may find a different release of the CD you have (or a entry with no associated disc id), of course it is also more likely to find entries in the database that have no relationship at all to your CD.
I don’t believe that Brennan currently implements the fuzzy search method hopefully this is something that can be implemented in the future. However although this would probably significantly improve the likelihood of finding the CD in the database it doesn’t solve everything. There are also entries in the Musicbrainz database that not only don’t have a disc id but also don’t have any information regarding track lengths and therefore it will be impossible to find them these entries automatically.
There are other mechanisms to search the database, such as a simple text search of album and artist names, but they would need to be implemented. The Musicbrainz database is a very different beast to the old FreeDB/CDDB database. Unfortunately the programming interface to the database only has limited documentation and a fare bit of reverse engineering is required to decode the responses from the database, also these responses can be huge and vary a lot leaving plenty of opportunities for bugs, so this is unlikely to be a insignificant amount of work.