Yes, if the licence allows re-distribution then yes you could put the paper in the several repositories as long as the licence under which the paper is made available there is compliant with the original licence.
One caveat; you generally shouldn't be publishing the paper initially or originally in two separate places. This is the Ingelfinger Rule. (Preprints are an exception to this of course.) Any additional locations you put the paper will likely at a minimum have to refer to the original place of publication, the version of record.
There are reasons you might wish to link to the original version rather than post copies. For example, having consolidated download statistics or other metrics is hard if there are multiple copies floating around.
This post has been migrated from the Open Science private beta at StackExchange (A51.SE)