Adding a response in case someone else has a similar issue. It's difficult to answer without an example, so let's say this would be selecting 2 of a possible 10 brands to be rated if respondent is familiar with a given brand. That's a pretty standard situation. For our purposes, the brands are referred to as B1, B2, ..., B10. As per OP's example, we need exclusions, so let's say B1, B2, and B3 are all made by the survey sponsor, and they want to make sure they get enough ratings for competitors. Thus a max of 1 chosen brands can come from the set of B1, B2, or B3.
OP is using what I'd consider in 2025 to be "old code" (i.e. standard code from before updates were made in how FORSTA functions) here. It can still work, but it's harder work than necessary, so my first suggestion is to just use quota objects and let them do what they're built for. It's perhaps inelegant, but quota objects are easy for the next programmer of unknown experience to follow. To that end, I would use two quota objects. By default, when a respondent qualifies for more than one quota cell, the cell with the lowest completion % towards that cell's limit is chosen. If no limits are set, it just goes by lowest count.
The first quota object (PICK_1) would choose one brand IF:
PAGE BREAK
The second quota object (PICK_2) would then choose one brand IF:
That would give you the two brands you needed.