What you dismiss as 'details' matter a great deal. Yes, the world is getting warmer, but how much warmer it will get is not merely some unimportant detail. If temperatures will increase by 0-2 degrees the impacts are easily dealt with given the time frames involved. If temperatures will increase by 2-5 degrees we should really look into cutting CO2 emissions now. If temperatures will increase by more than 5 degrees cutting CO2 emissions is pointless because the increase in CO2 we have already caused has already screwed us.
This is why the details matter, and why the things that we do not fully understand are important. Yes, it is settled science that CO2 is increasing and that in isolation, rising CO2 causes higher temperatures. However, CO2 is not rising in isolation, it is rising in the context of the entire planet with a whole host of other climate forcings that are also changing over time. The precise nature of how rising CO2 feeds back into other climate forcings (climate sensitivity) is immensely important and still unknown. Also, the fact that we do not fully understand the underlying natural cycles of climate makes it impossible to determine how much of the change we are seeing is the result of anthropogenic effects and how much is natural.
Knowing how much warmer the world will get is the driving force behind creating computer models. However, it is important to understand their limitations. Regarding computer models, John von Neumann once said "with four parameters I can fit an elephant, and with five I can make him wiggle his trunk." In other words, once you start dealing with 4+ parameters the number of possible solutions that will fit the data becomes astronomical and you have no way to verify which solution is the 'correct' one. When you consider that mathematical reality with the fact that IPCC Climate models use of 8-15 different forcing parameters to determine what climate will look like, you realize why considering the output of computer models as 'truth' is pure folly. The IPCC report section on computer models explicitly state that computer models are not useful as predictions tools, and yet people routinely say 'temperatures will rise by X degrees over the next century' as if it is a scientifically proven fact. It is not.
= new reply since forum marked as read