I think the most interesting thing that could be done with Strategic Ops is to improve their interactivity. With both XCOM's Spy Ops and XCOM 2's Resistance Ops, all you do with them is set the agent on their task then forget about them until their timer has run out. It's quite easy to forget that you've set someone on an Op because they're just a little reminder down in the corner along with all the other alerts. I've mentioned this previously but I'd like to do so again. There's a game (at least, there's a prototype of a game) called Net Gain. Now, Net Gain is all about progress bars which is not very interesting. To spice up the progress bars, Net Gain interrupts the progress bars with occasional crisis points. The player has to make a decision at these crisis points and the decision can govern the rest of the op. Here's an example:
This is something that's also used in Battletech. During the interstitial period flying between missions, occasional crises spring up which the player has to make a decision on.
I think this could be used to good effect in Xenonauts. You could tie a crisis decision to a timer, and give the player a short period of time to make an important decision. Let's take an example. Perhaps you've got an infiltration mission, and the crisis textbox throws up the following
"Commander, we have received an emergency communication from our agent. He believes he has been followed for the past several week by a shadowy figure, and is convinced it is an enemy agent stalking him. The agent has counter-tailed the figure and is in a position to take out the target. Permission to proceed?"
Then you might be given the choice to kill the figure (which may kill an enemy agent or might kill an innocent), pull out completely (ending the op), or ignore the target (potentially avoiding a messy scene, or imperiling the agent). If you tie it to a timer so the player has only a short period of time to make a key decision you introduce a degree of pressure on the decision. Furthermore, the information on the textbox might be more or less accurate. Using the prior example, the agent may be convinced or mostly certain, with every lesser degree of certainty giving the player clues how to proceed.
A useful sidenote is you can make all of this just text and it wouldn't loose the impact of a crisis, making it a lot cheaper to churn out crises.