I have just set up and run meme-chip to run locally for my eCLIP analysis.
I haven't tried HOMER.
I ran meme-chip vs. the human RBP motif dataset. You can get a gff file of the motifs and load that into IGV along with a bedgraph file of your peaks from CTK analysis.
You should further use bedtools intersect with that gff file and a genome features file to get genes containing motifs. I'll try this soon myself.
In my case, I was able to identify a motif that makes sense based on the biology of the protein and other research in the field, so I am somewhat confident in that motif.
Hope this is helpful, although it is not a direct answer.
Steve