When you're powering 2 cabinets, the sensitivity difference of the two is what determines the loudness difference - that's what you're hearing. No simple way to fix that. The question that remains is: are you getting something out of that combination?
You say that using 2 cabinets means things don't have to work as hard. If you have two equal sensitivity cabinets, that makes some sense, so long as the impedance of the combination is OK for the amp - the two speakers each work less hard, and you get a benefit from the increased efficiency due to mutual coupling of the speakers. If, however, the cabinets are of unequal sensitivity, the situation changes. If they're a little different (a few dB), things are still OK, even if one is producing a bit more than the other - you still get some benefit from using both,though not as much as if they're identical. If they're different enough in sensitivity (say 6 dB or more), then (assuming the more efficient cabinet can handle the power required to play loud enough) you're wasting amplifier power driving the less efficient one, and would be better off disconnecting it.
I have a couple cabinets that are different in driver size (gasp) and overall size. But, they're phase matched well enough that they work well together at low frequencies, and they're only a couple dB apart in sensitivity - they're designed so that I can use them together and get benefits from an "unmatched" pair of cabinets. With randomly chosen cabinets, you get...more random results. If you like the result, good. If not, well, they weren't designed from the beginning to work well together, and as a result....sometimes they don't.
