If I wanna prove V is a vector space, I need to find v, w and so on belongs to V, and then check properies one by one.
This is exactly the idea. However, you are not supposed to do this for the $V$ that appears in your problem. Your problem is asking you to verify that the collection of linear transformations $T:V\to W$ (with appropriate operations) is a vector space. Let us call this collection "$L$."
Using your words, we need to show that elements of this collection $L$ satisfy the definitions of a vector space. Let's take two elements $L$, and call them $T_1$ and $T_2$. Both are linear transformations $V \to W$. Verifying closure under linearity is basically showing that $T_1 + T_2$ is also in $L$, or in other words, $T_1+T_2$ is also a linear transformation $V \to W$, which is the "main idea" that you stated.
There is a question of what the addition in "$T_1+T_2$" means. This is your last sentence: by definition $T_1+T_2$ is the map defined by $(T_1+T_2)(x) = T_1(x)+T_2(x)$. All that is left to verify is that this map is also linear.
Response to comment:
\begin{align}(T_1+T_2)(av_1+bv_2) &= T_1(av_1+bv_2) + T_2(av_1+bv_2) \\&= aT_1(v_1) + bT_1(v_2) + aT_2(v_1)+bT_2(v_2)\\&=a(T_1+T_2)(v_1) + b(T_1+T_2)(v_2).\end{align}