You are right, in your set of outcomes only outcome 2 is Pareto optimal. This is a direct application of the definition: for outcome 1, player 2's position can be improved, without worsening player 1's position. Thus, outcome 1 is not Pareto optimal.
These are exactly the kinds of comparisons you do in economics. Suppose you give away some item, and two players have private information about how much they value the item, $\theta_1>0,\theta_2>0$. This is the "utility" they obtain when they receive the good. Suppose $\theta_1>\theta_2$. If transfers are allowed, it is only Pareto-optimal to give the good to the one with higher valuation, i.e., only giving it to $i=1$ is Pareto optimal. This is because with transfers, you can take away a transfer from $i=1$ and give it to $i=2$ such that $2$ is indifferent between getting the item (without transfers) and not getting it (being compensated by transfer), while player 1 strictly prefers getting the item with transfers. Hence, giving the item to player 2 is not Pareto optimal, because there is another allocation that makes player 1 bettor off, without making 2 worse off. (Note that this reasoning does not apply if transfers are not possible.)