Voting on a mon numerous times probably isn't as dangerous as it seems on paper (just look at the Excadrill votes since R1). But I definitely think nominations for these mons should be held to a higher standard and people should have to explain why, regardless of past votes for OU, that mon deserves to be voted on again. I definitely think it'd be bad to drop a blanket statement that something can't be voted on again if it was voted OU a few times, but at the same time I do agree that it shouldn't be voted on every round unless there's good reason to vote on it again (like franky said, "metagame shifts").
As for the second "flaw," I don't see how this is a problem. Generally, what's voted on would be voted on regardless of who does the nominating. Maybe if one round all the nominations from users who didn't qualify contained Pikachu, I'd see the problem. But even then, it wouldn't do any harm because the voters would undoubtedly not ban it (well maybe Pikachu is a bad example...so strong). Now in a banhappy community, I could see something like that being a problem. Where no matter what the general voter pool sees in front of them, they just put a ban next to it. However, this is clearly not the case in this community. Smith brings up when people bandwagoned on my Reuniclus nomination (which isn't really the case, but let's just go with it). That round, reuniclus received an absolutely pitiful vote (iirc, it received the least Uber votes). I'm personally a fan of allowing maximum community involvement in the suspect testing process. The best way to do that is to allow for nominations from everyone. As long as it does no tangible harm to the suspect testing process (which it hasn't/won't), I see no reason to restrict nominations to only qualified voters.