I agree with Hillel on the answer to the first question.
I also agree on his second answer. Just to expand: in TLA+, strings are sequences of characters, so "Hello" is semantically identical to a value that could be represented as <<'H', 'e', 'l', 'l', 'o'>> if we had a notation for characters [1]. This means that a string cannot be distinguished from a tuple. However, you can check if S \subseteq STRING or S \subseteq Seq(Int), for example.
Stephan
[1] Traditionally, TLC has handled strings natively without being fully faithful to the TLA+ semantics.