By the sound of it, these are all in house. The problem is that it may take (numbers out of my ass) 20 man-years to make a game, but there are constraints on how those man-years can be used. You can't throw 20 guys at it and be done in a year, since some parts have to wait for other things to be done first. It might take 2 guys a year and a half each to get the engine stable (since they're using a custom engine developed in-house), then the other programmers need a year or so to work on all the objects that need to work within that engine framework - but they can't start until the engine is relatively stable. Throw in the art assets (but you may not know what you need art *for* until other people have gotten to a certain point), etc. Add iteration steps (programming for one step finds a flaw that needs to be addressed elsewhere) and the delays add up.
The point of that (which is probably a painfully bad description, since I can barely program my microwave's clock) is that the engine programmers are only working on the project part time - after they've passed off to the other teams, they can spend their time doing the groundwork for the next project. Same with the AI programmers, artists, etc. I wouldn't be surprised if the GalCiv 3 project started 3 years before they mentioned anything publicly.
As a reference point, take the World of Warcraft team. Blizzard has said they have people working on the *next* expansion, when the current expansion under development is still 6-9 months from release. It's a good bet that next expansion won't even be publicly announced for 2 years or more.
It's true. Some tasks in developement can be easiily done in parallel and throwing more people at it will get you there faster. Other tasks are not, and this famous quote applies:
"adding manpower to a late software project makes it later" - The Mythical Man Month
People who are doing something like basic design or prototyping for a future game are not necessarily people that would be of any use in the Galciv project as it currently stands, as that work was already done. Writers are like that. If the storyline and dialog is already written, what is a writer going to do now? Better to put them where they can be useful.
There's other cases where B can't be started until A is finished, and you don't want the team that does B twiddling their thumbs doing nothing waiting for A.
Incidentally, this is also one of the major causes of day 1 DLC. What happens is when a game gets closer to release, more and more people that were on the project have nothing to do. Content producers, artists, and the like have no value once the game goes into the final testing phase, as it has to be *done* by that point so it can be prepared for release & certification (on consoles). So do you lay them off, put them on another project, or send them to work making DLC instead?