Jefferson-
If you can use array_filter, that is easiest, and you can do it in 1.91+. If you know the products you're asking about, you can have subjects click which they use, then have follow-up questions that use array_filter to just show follow-up questions relevant to the selected products.
If you need to ask questions that can't be array filtered, here is how you could do that in 1.92RC1:
Say you use a multiple short text question type to let people enter the product names. Let's call this question "Products", and the codes for them #s A-Z
Then, say you have 5 follow-up questions about each product. Let's call them FU_A1, FU_A2, ..., FU_A5
...
FU_Z1, ... FU_Z5
The easiest way is to put the 5 follow-up questions in its own group. That way you can export the group, re-import it, and change the question names quickly and easily.
Soon, you'll be able to use group-level relevance. So, for the group, use a Relevance equation like
for group FU_A (which contains questions FU_A1 - FU_A5).
For now, however, you need to assign relevance to each question within the group. So, you'd put the same equation for questions FU_A1 - FU_A5.
When you run the survey, groups that have no relevant questions are completely skipped.
I'd also recommend running with the Question Index on. That way, if people want to go back and change lists of products, they can click back to the end, then click on a later question, and LimeSurvey will ensure that any newly mandatory questions are asked (and that any newly irrelevant questions are NULLed out).
/Tom