This sort of use of the dot as punctuation can be just thought of as marking a pause, so that
∀x ( x ∈ {a,b} ↔. x=a ∨ x=b )
is in effect
∀x ( x ∈ {a,b} ↔ [wait for it ...!] x=a ∨ x=b )
so naturally gathering what follows into a unit. Hence it is a bracketing device. (And "↔." is no more a unit with its own significance than is, say, "↔ ("
The once-common use of dots for bracketing duty seems to have its origins in Principia Mathematica. I'd forgotten I'd written that answer on official dotty conventions at https://math.stackexchange.com/q/312074, so thanks for the link that saves me writing something similar! :) But as I say, in relaxed contexts like the present example, you needn't worry about official conventions. Just think of it as like a pause to break up a sentence and group it. As we do in ordinary language. Thus compare the familiar example of the party invitation ...
Bring your partner or come alone [pause] and have a good time
Bring your partner [pause] or come alone and have a good time
Without the spoken pause, or the written comma, it would be ambiguous. Likewise without the dot, or bracketing, or an operator-precedence-convention,
∀x ( x ∈ {a,b} ↔ x=a ∨ x=b )
would be potentially ambiguous (even though one reading is daft).