The semantic correlation "skin":"colour" is observed also in several other roots, thus the etymology is fully justified. The phonetic side of it is also quite regular (in PWC we should expect *L - a phoneme which lacks as such, and must have merged early with *Ĺ). In WC languages the root is verbal, in EC - nominal; the only trace of its verbal usage is perhaps a Lak. form (recorded in the Khosrekh dialect) =u=k:a-n 'to paint'.
Abdokov (1983, 77) quotes also Kab. La, q:ʷǝ-La 'inner side of skin' (which he compares with heterogeneous EC material, having nothing to do with the present root); if it really exists (we could not find it in existing dictionaries), it may be a remnant of *Ł_ŏli with the meaning 'skin'.