In my experience, most cigars improve with age. Though it seems to me that the more aged the tobacco, the less improvement you'd see with additional aging--that would make sense, anyway.
With most NCs, a little extra aging will allow harshness to mellow out (too much "char" taste, ammonia, etc.) and flavors to coalesce. Sometimes the flavors become more pronounced, sometimes a creamy element come in that wasn't in before, and of course any cigar will get smoother with age. [NOTE--sometimes you might NOT want a cigar to smooth out, such as with a ligero stick that has a kick that you really like (Tat T110 maybe)].
Many manufacturers ship their NC cigars sooner than I'd like. Tats, for example, usually need about 6 months to "calm down" imvho. A Noella with a box date of Jan 2010 will not be as enjoyed by your's truly as much as one with a box date of June 2009, for example. Tats are usually shipped kinda young to me.
I had a Cubao toro with about 1.5 years in the humi and thought it was a LOT better than they are fresh (though I like them fresh as well). There was a caramel and vanilla type flavor in it that was just awesome, yet it didn't loose it's peppery goodness!
The issue of whether or not a manufacturer should ship their cigars so that they're smokable out of the box or not is another topic altogether. The issue at hand is one of cost (storing cigars for years would add a significant price to the stick--i.e. Padron Anniversary). Pete doesn't age his tobacco extensively before it's rolled (in his opinion, everyone likes different things, so he releases it when he does knowing people always have the option of aging them). I do like the Tat red label right off the truck, but a fresh Angeles tastes a LOT different than one with 2 years in the humi (seriously--do a comparison, it's really cool!).
But to each his own. I've heard some friends say that Padron regular lines are MUCH better with age (I've found that to be true as well), but the Anniversary blends can loose a little something with too much age. I'm not likely to have any on hand long enough to test that particular theory
