Well, maybe we disagree, maybe I think gameplay is more important than a coherent style, or maybe I am just more pragmatic

I only see those existing assets as a base to work with and improve on, call them place-holders or less then perfect solutions until a more polished and unique one is made. However quite honestly, the latter is wishful thinking as most likely they will stay in a loooong time; still better than having an unplayable prototype for years though.
Obviously this doesn't apply to all projects, and going the "retro" way to vastly speed up development of art assets is also a viable strategy, but if you want to aim for somewhat modern graphics, I see no way around reusing existing assets at least for some time.
The lack of playablility is IMHO the main reason why projects fail... a single developer will only stay for that long dedicated to a project, thus if he or she hasn't managed to attract a community by then to continue the project, it is a dead fish in the water (as so many FOSS game projects). Thus try to make it fun and playable as soon as possible!