Errrrr... I'm going to be as nice about this as I can:
The FL executable was compiled, like any other application. Decompiling it gives you source code that you can then (theoretically speaking) recompile into a working executable if:
1. You knew exactly what compiler the FL team used, and what settings they enabled. Insofar as I know, we don't even know that.
2. You were aware of, and counter-acted, any obfuscation methods used when they compiled it. Very tricky, if we don't know #1- we need to sort the garbage from the useful information.
3. You could then figure out what variables were called, and what subroutines were doing what, etc., etc., etc., because this information is stripped from the code when it's compiled.
In short, decompiling an executable and rebuilding the sourcecode into something vaguely like what modders would need before we could make any headway into adding new features... would be very, very hard.
It would be easier, in most ways, to simply (ha ha... "simply"
write a new game engine from scratch that could use FL's game content natively, read its scripting, models etc., and do all of the things that the FL game engine does, but without the restrictions imposed by not having the source (i.e., the project would be open-source so that modders could write their own engine modifications). It would even be legal to distribute this game engine, so long as it didn't contain any content produced by Digital Anvil or Microsoft.
There *are* precedents for this. The Total Annihilation community, after 6 years, produced a game engine called TA Spring, which is nearing release. Spring supports all of TA's engine features, supports the native file format for game objects (3DO) and also supports TA's native scripting languages, which run all of the game objects. It's written to support OpenGL, has truely 3D maps, and supports a bunch of gee-whiz lighting effects and other things.
It took a team of expert coders nearly 2 years, using all of their spare time, and it's still not out yet (although it shows no signs of being vaporware or collapsing). For a game engine that's as advanced as Freelancer's is (and in many ways, it's quite advanced, although it's not top-shelf anymore) it'd be very hard to duplicate the engine, support the existing content, etc., etc. etc.
I would like to see people thinking about making an FL engine clone that didn't support all of the advanced features (such as in-model LODs, which quite honestly don't seem to matter much in terms of game performance), but I doubt anybody has the energy, talent and enthusiasm to try to hack something like this. And then even if they did, they'd still have the AI coding (another big job), server/client coding, and the new code to support user requests (which is an incredibly long wishlist, even if I were the sole dictator in charge of implementation).
Soooo... basically, it doesn't matter if you are contemplating something illegal, sir- what you're asking for is extremely difficult no matter what laws are broken