My motivation is the bad implementation of intelligence in civ like games (with the exception of Master of Orion). At least that's my point of view.
'My' intelligence model would reduce micromanagement and add some new elements to strategic gameplay. To be honest, 'my' intelligence model is a conglomerate of what I've read so far, seen in other games and some ideas of my own.

No spies?
There won't be spy units visible on the map. Every civilization has an intelligence agency which must be funded from treasury. It is divided into internal security (isec) and departments for each civ. The player can decide which departments get certain amounts of money. Here is an example how it works:
Imagine you are the Romans and you have 100 CC to spend every turn. Then there are the Carthaginians and Macedonians, too. Of course you want to know about their plans. You decide to put 50 CC into isec and because of some special reason you give 30 CC to the Carthaginian department and 20 CC to the Macedonian department.
That's the way it works in Master of Orion, nothing new so far.

What does intelligence do with your money?
After adjusting the fundings the player need not build any spies. They are hired by the intelligence agency. The maximum number of spies in each department including isec depends on the funding. Spy activities grow slowly so there has to be a certain time before the maximum number of spies is reached. Losses should also be taken into account. Once a spy network (at least one spy, numbers are abstract) has been established in a foreign civ or as isec the spies will send their reports to the capital.
Also not very innovative, but let's see.

Establishing spy networks and reports
Spies can always get caught or killed by enemy agents. So while operating in a foreign civ the enemy isec has a chance of catching/killing each spy. This chance must be calculated based on the number of spies in isec. An isec agent can also be 'removed' by enemy spies operating in his country.
The following list contains the information that may be obtained by spy reports. It is not complete.

Reports on foreign countries

tech status

demographics

happiness of population

whereabouts of army groups

planned movement of army groups

locations of important weapon factories

diplomatic plans

Reports on internal security

terrorism

detected operations of enemy spies

loyalty of governors

danger of civil war or revolution

Special operations!
Having all this information about the opponents is very nice, but who cares? Therefore special operations are a must but they should be expensive and rare. That means the player has to pay extra money for the extra action. And it will take time so enemy isec can torpedize it.
Perhaps it would be fun and less frustrating having more results than success or failure.
Here's the -of course uncomplete- list with explanations and possible outcomes:

Steal technology

Stealing the technology only makes it cheaper to research by yourself. A minor success results in a marginal cost reduction. Corrupted data might even slow down research on the stolen subject.

Incite revolution

If one class of the opponent's society is unhappy spies provide them with weapons so that they can revolt properly. The success determines the military strength of the revolutionaries. Perhaps there will only be a period of terrorism. An alternative would be the support of separatist province leaders which would cause a declaration of independence.

Fund guerillas

Support one side in a civil war. The enemy isec might get the money if you try to help the opposition. Supply guerillas and partisans in territories lost to the enemy in a war. A major success could be the withdrawal of enemey forces caused by the destruction of their supply lines.

Bribe governor

This results in more accurate and up to date information about the civilization. Slight chance that the governor is prepared by isec. Then all data will be corrupt.

Infiltrate enemy intelligence

Get all their spy reports until isec finds out.

Thanks for the writup. I think this is a good set of ideas that is a very solid start on intelligence issues. I've only got mostly minor things to add.

Buying spies should be more difficult for placing them in insular states like the Soviet Union. And easier in relatively open states. Costs of spies should also change with tech level and relative tech level between the two sides IMO.

Special Ops:

All the examples you sighted should have counter-measures from the other side applicable. FE the steal technology should have a counter-op called secrecy where certain techs are especially well protected, etc.

should also include suppress dissidents within the civ.

Management:

Also, in keeping with the Hands-Off management style available in Clash, you should be able to tell the AI for your Intelligence service to take this 100CC and hurt the Brobdignagians in the most economical way, and not have to worry about the details. Or have the Intelligence Commisar present you with "his" ideas and approve or cancel them.

Please add further suggestions to the "Diplomacy Model v1.1" thread.

This one was bumped for Kull, duke of diplomacy.