You accuse the current name of ambiguity; you then imply by your choice
of new name that the protocol specified in RFC 1421 might be used to
support authentication of a "memo", or "document", generally.
Document authentication implies that a signer signed the document
content, possibly with signatory witnesses, with a thousand legal
consequences. I do not believe RFC 1422 offers that security service,
offering (based on the text); only "message origin" authentication
(possibly with non-repudiation). Should you decide that RFC 1421 does
offer the content authentication security service, then you should
change the specification of services services PEM (RFC 1421) offers,
also.
If you drop the messaging from PEM (1421 protocol), then you change the
communication semantics, and the whole basis for the assurances
attached to the purported security of that (non-)communication.
Message Stores do need protection - provided generally from a trusted
operating system, as in Oracle's security products. Its a misnomer to
assert that secure messaging translates to secure message
storing/retrival/listing etc. just because you change the name of the
messaging protocol.