We know that ZFC was formulated to avoid some paradoxes inherent to Cantor's naive set theory, such as Russell's paradox, which inquires about the truth of the existence of the set of all sets.
The notion of set, however, comes prior to any axiomatic set theory: to write ZFC's nine axioms, we are using sets of variables, constants, functions and relations. Those primitive, pre-ZFC sets, don't generate paradoxes simply because we choose never to write paradoxes with them: those sets are solely used to generate ZFC.
The point is: can we avoid ZFC in its entirety by keeping on formulating mathematics with those primitive sets and never touching any paradoxes?