Intelligence and Espionage were very limited and cheezy in HOI3, especially with the later expansions. The effects of several missions were heavily toned down to take into account the potential for multi-player mass spamming of spies on a target, so each country's espionage program now has relatively little individual impact.
What I'd prefer instead of "tech stealing" would be for spies to more frequently "steal" variable amounts of points of research toward those techs, instead of stealing an entire tech increase "lock, stock, and barrel" at long intervals. Realistically, any technological theft would still need to be "researched" to modify it for the different manufacturing capabilities of the country which acquired it. It could also include some aspects of "reverse engineering", where you gain some points by capturing enemy equipment with different tech focuses than your own. The end result would be somewhat closer to the "neighbor" effect in EU3, where you got a small but significant research bonus from each neighboring country with higher techs, except that in this case it would be applied in random chunks to various individual items.
The ability to support different parties than your own, to sway an election between two close rivals, would be helpful. As it is, the fixed rate of increase per spy to boost a rival party in a larger country versus a smaller one makes it far too easy to bring about significant changes in the countries least likely to change, while not allowing those countries which were easily influenced historically to be swayed. Perhaps the effect should be based on the target country's IC or population, modified by National Unity, so small countries with internal divisions would be affected more easily than larger monolithic countries. Dividing up the effects and susceptibilities into external pressure and internal unrest might represent different national situations better than a single value.
For example, GER should be large enough to be difficult to affect externally, but internally disunited to some degree, so the German player would need to concern himself (or herself) with solidifying power, and less about foreign agents overthrowing the government. A small but internally stable country like Luxemburg or Portugal would be easier to sway from outside on account of size, but harder to change internally on account of unity, and a fairly small fragmented country like Yugoslavia or Spain would be subject to both external and internal pressures.
