In a money economy, each good has a price expressed in terms of dollars. But what's important is not the price itself, what's important is relative prices, the price of one good in terms of another. If milk has a price of $4 and gasoline has the same price, $4, then we know that buying one gallon of gasoline means giving up one gallon of milk. If purchasing milk is the highest valued alternative to purchasing gasoline, then this measures the cost of the purchase. Thus, the cost of a good is what you give up to get it, and this is measured as a relative price.
Relative prices provide important signals to both producers and consumers. They give consumers a means of deciding how to optimally allocate their purchases, and they direct the flow of resources to their most profitable locations. Because of this, it's important that price signals be as accurate as possible.
When there are sluggishly adjusting prices in an economy, accurate signaling becomes a problem since some relative prices will slip out of alignment with their optimal values, and this causes a misdirection of resources across sectors. The wrong mix of goods is produced and consumed, and the outcome is suboptimal.
Inflation can make this worse because it accelerates the rate at which relative prices fall out of line with optimal values. If all prices are perfectly flexible and adjust instantaneously in response to any shock, then inflation isn't a problem since relative prices will always be maintained at their proper values. But if prices aren't perfectly flexible, or there are other rigidities, then the Fed has to worry about keeping inflation from distorting relative prices and causing problems.
So one job of the Fed is to make sure that the price signals in the economy are as pure as possible, and that is the point behind inflation targeting. It's a way of trying to maximize the information content of relative prices so that resources are directed, as much as possible, to where they would go under ideal circumstances.
But think about commodity prices, oil perhaps. Is the run up in prices a true signal of scarcity? Is the price increase accurately reflecting underlying conditions? If so, then this is not the kind of a change in prices that the Fed should try to mute even if there is a temporary increase in the inflation rate because of it. It's signaling to substitute away, to bring more supply to market, to innovate, to drive less, etc. It's a change in relative prices that is providing an important signal to consumers and producers.
But if the increase is due to speculation, that's a different matter. In that case, the increase in price is providing a false signal to market participants and that distorts resource flows. Here the Fed would, if it could, want to try to temper the increase in prices and institute other changes as needed to allow fundamentals to take over once again.
The intense debate right now about whether the increase in oil prices and commodity prices more generally is due to fundamentals or due to speculation that is driving prices away from fundamentals - i.e. whether or not there's a bubble - shows how hard it is for the Fed to identify and do something about bubbles as they are inflating. Do we have a bubble in commodity markets or not? Is it the Fed's fault? Are prices going up partly from fundamentals and partly from speculation? Then how much of each? (I think some people use the word speculation when what they really mean is that oil producers have excessive market power that allows them to increase prices almost at will. Market failure can also cause distorted prices, but except for financial market regulation, this is outside the Fed's purview, so, while it's important to maintain competitive markets, that's for another discussion.)
We can't always have zero (economic) profit in every market at every point in time. In order for resources to be attracted to their best uses in a dynamic economy, there have to be profits - there has to be a pie to fight over - and sometimes that pie will be huge if there has been a large shock to a particular market and there will be lots of money to be made. That's not a bubble, that's the market doing what it's supposed to do and we don't want to get in the way of the market's ability to move resources where they are needed the most (though there may be reasons to slow the adjustment down at times, but that's another discussion as well). It will look a lot like a bubble when there is a large amount of excess supply or demand, but popping it would be a mistake.
Thus, while I certainly think would certainly be desirable for the Fed to pop speculative bubbles, practically it will be hard to do and mistakes can be costly. If, somehow, they are relatively certain that a bubble exists, action is warranted, but this is likely to be a rare event.