While I'm at it Karen let me add that the models as far as I know do not have the capability of modeling the atmosphere on a mesoscale level. You keep hammering the mesoscale part, and that is true. However when we forecast using the models we look at other things more on a synoptic scale. While it may not necessarily model a mesoscale feature it may model some trends that could result in mesoscale features. For instance if it models precipitation from the night before these typically can result in an outflow boundary which can then be analyzed to possibly determine their potential part in the storm environment. Of course the models usually don't do that good with precip or precip feedback either. When I look at the models I try and look at the big picture, see what players there are and what trends are setting up. At that point you can look at satellite, and surface data to see what is truly inbound. Morning soundings will give you an idea of what type of environment is where and then project where it will be. So first you look at a 'Macro' scale for the environment you want to 'play in' and then you look at real data to see what trends may be taking place. I liked what I saw the night before and left. Along the way based on what I see developing I can modify my forecast for what is transpiring. One thing I would look for is an outflow boundary, cu development, even a meso low developing. Perhaps it takes a lot to technically identify a meso low, but usually using surface data and a tool such as Digital Atmosphere it will easily show you them developing especially if you look at altimeter settings and not sea level pressure. I've seen them numerous times this way. Whenever I see a small low with streamlines starting to wrap around it and winds backing around it, and convergence maximizing near it I know that whether it is technically a proven meso low or not doesn't matter. What matters is the fact that it will influence weather and in many cases especially if it occurs in the region of my tornado forecast target will be the go to spot for tornadoes.
In summary with regard to mesolows, outflow boundaries, and current forecasting - I don't think you have to see 'every tree' to 'view the forest and determine that it is on fire'.