Big or small

I am often confused by the dilemma of big vs small. The dilemma arises often in the context of social change, research, office work and in daily routines. The big vs small debate is stated as follows:

You have lots of work to do. Work comprises of big tasks and small tasks. Big tasks means bigger implications and small tasks mean not-so-big implications. Now which one should be done first. Bigger task would mean that you are tackling the main problem first while working on the smaller on would mean that you can get over it quickly and focus on main tasks. Which one should be done first??
Clearly a round-robin algorithm is one of the worst one can think of, i am not an OS..i am human. Weighted round-robin (where time quanta is proportional to task size) may work for some but not for me.
Another aspect of this problem is that many argue that there is no big task, only small ones...if you see a big one, you lack modularity.....I wonder if this rule is a general rule.
Psychological aspect says that prolonged attachment to a single problem may lead to de-motivation.

Is there a general rule to solve this problem???