Peregrin Took’s answer is completely correct: the list of options you have for your bonus feats change, and you no longer have access to the previous list, but those lists were developed extremely early in 3.5e and can and should be expanded by the DM as necessary.
But I want to get a little bit more into the situation here with respect to combining those variants with other variants, because that gets a bit more complex.
At the time that Dragon vol. 310 was published (August 2003), the entire concept of “variant” anything didn’t really exist. The 3.5e Dungeon Master’s Guide discusses how DMs could vary things in their own games, and had some ideas and examples of the ways in which the game could vary, but these weren’t pitched as “content” per se, just as ideas. Unearthed Arcana would later (February 2004) catalog some variants and put some rules around them, but even then we have a whole lot of unanswered questions about these variants.
And the reason for all of this is that these variants are still largely following in the footsteps of the Dungeon Master’s Guide suggestions: these are ideas, largely for the DM, to flesh out their game world. They necessarily require massaging into place. For example, Unearthed Arcana has some very tentative suggestions for multiclassing between a class and its own variant, and how to handle class features that both get. It doesn’t have rules for multiclassing between a class and a variant of another class that gets the same features. Dragon vol. 310 is in the same boat, and if anything, less detailed.
So combining the features of two different fighter variants isn’t terribly clear. Unearthed Arcana explicitly states that “this [fighter] variant can also be combined with the thug variant [fighter],” even though both of those variants lose the bonus feat at 1st level—and doesn’t specify whether such a fighter should lose sneak attack at 1st level to reflect the lost bonus feat at 1st level. For that matter, when the sneak-attack variant fighter gets “sneak attack (as rogue)” instead of “bonus feats,” it’s not clear if they should get the exact same sneak attack progression as the rogue (+1d6 each odd level), or instead get a progression that mirrors the bonus feats they gave up (each even level as well as 1st). The latter means the sneak-attack fighter has more sneak attack damage than a rogue—maybe OK because they’re more of a combatant, maybe? It’s unclear.
And there are myriad more questions besides. We know the sneak-attack variant fighter can also be a thug because Unearthed Arcana explicitly says so, but it doesn’t otherwise address combining variants.
The general consensus—made official by the same Paizo that published Dragon vol. 310 for their Pathfinder spin-off of 3.5e—is that as long as the two variants don’t replace or modify the same feature, they should freely combine. But the only explicit example we’ve got—the sneak-attack fighter and thug—seems to violate that rule. And, for example, if a feature replaces a fighter’s bonus feat with something else—such as that overpowering attack variant—can you still take it if the feat list you’re getting bonus feats from is different from the default fighter list, as it is for pugilists? We don’t know.
In short, variants always need massaging to get into the game, so as a player you should work pretty closely with your DM to figure out how everything works. We cannot really give you an answer here, only your DM can (and then only for that particular game).
All that said, for what it’s worth, I have an enormous amount of experience playing in and running games that are extremely free with variant combining. I’ve even allowed “reverse variants.” And I have basically never had any problems (the reverse variants a couple times, maybe). So should your DM be looking for expert opinions on the subject, mine would be to allow it all.
