In addition to language codes, there are also script codes. 4-letter ISO 15924 codes are incorporated into the Language Subtag Registry for IETF language tags. With these codes one could differentiate, for example, between Serbian written in the Cyrillic (sr-Cyrl) or Latin (sr-Latn) script, or mark romanized text as such.
The list of script codes can be found here: http://unicode.org/iso15924/iso15924-codes.html
By supporting this list, it would be possible to use cu-Cyrl and cu-Glag as language variants in Lexemes/Forms, which at the moment is not possible.