[guided]The goal of this step is to upgrade an $L^2$ estimate into a pointwise estimate on compact subsets. This is where holomorphicity is used.
Fix a compact set $K \subset \Omega$. Because $K$ is compact and $\Omega$ is open, there is a radius $r_K>0$ such that every closed ball of radius $r_K$ centered at a point of $K$ remains inside $\Omega$:
\begin{align*}
\overline{B}(z,r_K) \subset \Omega
\end{align*}
for every $z \in K$. Let $\alpha_{2n}:=\mathcal{L}^{2n}(B(0,1))$ be the Euclidean volume of the unit ball in $\mathbb{R}^{2n}$, so
\begin{align*}
\mathcal{L}^{2n}(B(z,r_K))=\alpha_{2n}r_K^{2n}.
\end{align*}
For indices $j,\ell \in \mathbb{N}$, define the difference function
\begin{align*}
h_{j\ell}:\Omega &\to \mathbb{C} \\
z &\mapsto f_j(z)-f_\ell(z).
\end{align*}
Since $f_j$ and $f_\ell$ are holomorphic on $\Omega$, their difference $h_{j\ell}$ is holomorphic on $\Omega$. The mean-value inequality for holomorphic functions says that, for each $z \in K$,
\begin{align*}
|h_{j\ell}(z)|^2
&\leq \frac{1}{\mathcal{L}^{2n}(B(z,r_K))}
\int_{B(z,r_K)} |h_{j\ell}(w)|^2 \, dV(w) \\
&= \frac{1}{\alpha_{2n}r_K^{2n}}
\int_{B(z,r_K)} |h_{j\ell}(w)|^2 \, dV(w).
\end{align*}
The ball $B(z,r_K)$ is contained in $\Omega$, and the integrand $|h_{j\ell}|^2$ is non-negative. Enlarging the integration domain from $B(z,r_K)$ to $\Omega$ therefore gives
\begin{align*}
|h_{j\ell}(z)|^2
\leq
\frac{1}{\alpha_{2n}r_K^{2n}}
\int_{\Omega} |h_{j\ell}(w)|^2 \, dV(w).
\end{align*}
Taking square roots and then the supremum over $z\in K$ yields
\begin{align*}
\sup_{z\in K}|f_j(z)-f_\ell(z)|
\leq C_K\|f_j-f_\ell\|_{L^2(\Omega)},
\qquad
C_K := \left(\alpha_{2n}r_K^{2n}\right)^{-1/2}.
\end{align*}
Since $f_j \to f$ in $L^2(\Omega,dV)$, the sequence $(f_j)$ is Cauchy in $L^2(\Omega,dV)$. The displayed estimate transfers this Cauchy property to the uniform norm on $K$. Thus $(f_j)$ is uniformly Cauchy on every compact subset of $\Omega$.[/guided]