The base case is clear, since if $n\gt 4, 5!=120\gt 25=5^2$
So assume $n=k$ which shows that $k!\gt k^2$.
Then if $n=k+1$,
$$(k+1)!=(k+1)k!$$ $$\gt (k+1)k^2 $$ Induction argument $$=k^3+k^2$$ $$\gt k^2+2k+1$$ $k^3>2k+1, \forall k\ge4$ $$=(k+1)^2$$
Is this a good proof? I never attempted the factorial induction proofs, and I first tried proving the other direction, that $n^2<n!$, and starting with $(k+1)^2$.
(Judging from your title, in which your result is $(n+1)!>n^2$) Your Induction Hypothesis should be $k!>(k-1)^2$, and thus you should be trying to prove $(k+1)!>k^2$, and you shouldn't bring in an inequality like $k^3>2k+1$ for $k\geq 4$ unless you've already proven it, so just make sure you prove that as a lemma somewhere in your proof.
But whatever the case, note that $(n+1)!=(n+1)n(n-1)!$. $(n+1)n=n^2+n>n^2$ for every $n>0$, so the result holds for more than just $n\geq 4$, it holds for $n\geq 1$.