To be honest I already find the current one cool enough, as a blend of both eastern and western mecha, they remind me of Xenoblade Chronicles X' mechas, but yours feel much more reasonable (arms/legs-size-wise), not to mention, you seem to have already put much work, thus hours, thus MONEY into it.
The Mk.II seems more in-line with eastern mechas, with human-like arms and legs being the giveaway, so you should think about the crowd you want to meet with this in mind, eastern (which means mostly Japan, since China should still be rather green on this matter so you have more of a leeway) or western?
Well, you could do both, with time, that is, since you'll probably want to monetize on cosmetics anyway I think it's bound to happen no matter what (to implement both styles, that is).
Usually, when developing games the most important thing is to bring out a fleshed out player controller that feels right as soon as possible, because it allows testing for: gameplay, combat, animations, systems implementation, skills usage, not to mention the morale impact on the developing team, it's a massive boost, to see your thing in action, ready for the world to come (quite literally).
Sure, the artstyle can be important in determining the graphics' style you want to run, but we're not arguing cell-shading over super-realistic so...
I think you should stick with the first one, that's your priority and you can't go wrong with priority driven development.
On a note I found this just the other day while scouring the internet for mecha images to get some inspiration:
https://www.kickstarter.com/projects/1609568567/code-hardcore-the-coolest-2d-mecha-battle-game
In the video they mentioned how they considered both eastern and western-styled mechas, so, heck, just contact'em to ask stuff, that can't hurt right? They're Chinese so you may want to check up on them to clear your doubts about that market.