I don't know the answer, but I have an answer in the case ##x## belongs to finite set of rationals ##S = \{ x_1,...,x_p\} \subset \mathbb{Q}##. It may or may not be useful for what you want to prove.
## \{f_n(x_1) \}## is a real bounded sequence so there is a converging subsequence ## \{f_{\sigma_1(n)}(x_1) \}## that converges to ##f(x_1)##
## \{f_{\sigma_1(n)}(x_2) \}## is a real bounded sequence so there is a converging subsequence ## \{f_{(\sigma_2 \circ \sigma_1)(n)}(x_2) \}## that converges to ##f(x_2)##. Furthermore ## \{f_{(\sigma_2 \circ \sigma_1)(n)}(x_1) \}## converges to ##f(x_1)## as a subsequence of ## \{f_{\sigma_1(n)}(x_1) \}##.
Repeating this process, you have ##\{ f_{(\sigma_p\circ ... \circ \sigma_1)(n)}(x_i)\} ## convergerges to ##f(x_i)## for all ##i = 1... p##.