Science on these is comparatively solid relative to many other public health movements, though its important to note that very, very little of it is experimental and there is still a lot we don't know. Regardless, we pretty much know that our current warnings are completely ineffective. We do have a lot of large-scale, correlation studies showing that these may be more effective. RE: cupcakes and alcohol, it may seem a relatively trivial distinction to many, but I do firmly believe the critical distinction is that there is no "safe" or "responsible" cigarette use...even one is bad.
We're trying to figure out what this means for ongoing studies, because we expose smokers to their cigarette packs in a few ongoing studies explicitly to evoke cravings and we aren't sure what (if any) effect these changes will have on acute effects. I've only seen one study on that to date, and it had weird/mixed findings that may or may not replicate.
The fact that they got the quitline number on there was big and critical - it implies this is productive.
I'm certainly biased since this is my area, but I have to disagree with the nanny state argument. Of course, I'm also not exactly against nanny states on issues like this for largely economic reasons. That said, they are not telling people they can or cannot smoke, nor does it affect contents of the actual product (though that is obviously being discussed as well). Those are all options being discussed, so this 1) seems relatively tame by comparison, 2) Helps us catch up with what countries with better healthcare are doing and 3) Seems a fair counterpoint to corporate dishonesty. This is getting a bit political but while "true" freedom may carry some long-term advantages, there is likely a whole lot of short-term suffering needed to get there. I think the government has a definite role in keeping corporate greed in check, and that a society that refuses to do so in the interest of being "Free" from the government (and enslaved to the dollar) is not a society I want any part in.