I don't think the MEME Suite has a tool that does exactly what you seem to be asking for.
SpaMo seems to be the closet to what you are looking for. SpaMo identifies the most common spacing between a primary motif and a secondary motif. It can consider multiple secondary motifs in one run.
Other tools that might be relevant are:
MAST identifies sequences that contain good matches to all members of a set of motifs.
MCAST scans sequences for clusters of co-occuring motifs.
GLAM2 performs de novo discovery of motifs that contain gaps
GLAM2Scan scans sequences for gapped motifs identified by GLAM2
But I don't think any of those are really what you are looking for.