Apologies in advance to anyone who knows about nuclear plants and wants to chime in. Here's what I think I've read:
Things at Fukushima Daiichi Unit Number 2 have apparently deteriorated. Reporting indicates an explosion occurred in the suppression pool (or "torus", a doughnut-shaped compartment is normally partially filled with water and is located below the nuclear core). The higher radiation readings seen around the plant after that explosion have increased fears that the containment vessel may have been breached (the suppression pool is connected to core containment vessel).
The other reactors which were the major concern yesterday seem to be cooling well at this point. There was a small fire in an ancillary building associated with a different reactor, but that reactor was shut down before the earthquake/tsunami, and the fire was a non-event.
It will be interesting to see what the news calls these problems at the Japanese nuclear facilities. I have seen references to the "accidents" at Fukushima and the other plants. As another commentator said " There are four entire trains missing and possibly washed away--will these events be called railroad accidents?". The backup generators needed to power the cooling water pumps (in the event the plant's own power was lost AND power from the grid was unavailable) were protected by walls that protected against a 10 meter wall of water, but this wave was at least 1/2 meter taller than that.
It is incredibly difficult to figure out what is actually happening at Fukushima by reading news reports. The nuclear industry uses very precise terminology, because it is very important to be precise about what you are describing. Journalists are not so precise. In the first instance, journalists are not trained in this area, don't know basic physics and engineering concepts and can't properly explain them. Accordingly, they conflate concepts and use the wrong terms. Almost as important is the journalists' overwhelming urge to sensationalize.
As I mentioned earlier, the key to avoiding a melt down of the fuel rods is to keep them covered with water inside the reactor pressure vessel. As the name suggests, the pressure inside is high. Pumping water into the vessel at a high back pressure is difficult if you need high volumes. To use the low pressure, high volume pumps, they need to depressurize the vessel. The outflow from the depressurization system is directed into the suppression pool, which is a water filled pool located in the torus. It scrubs the steam that exits the reactor pressure vessel and helps to remove some of the contamination. It also condenses the steam and thereby helps keep the pressure in the concrete containment dome from getting too high.
From what I have read, there was an explosion in the torus of Unit 2. It was probably caused by hydrogen gas released to the torus when the reactor vessel was depressurized. (recall that if the zirconium fuel cladding gets too hot, it reacts with any water to generate hydrogen gas.) It appears that the pressure reading on the torus then indicated atmospheric pressure, which could be an indication of a breach of the torus or containment dome (both of which are made of steel-lined concrete 6 to 8 feet thick). Or it could be a problem with the pressure gage. A visual inspection apparently failed to find any breach of the containment.
One critical fact that the news reports get wrong is that the torus is not connected to the reactor pressure vessel. The reactor pressure vessel is a big, ultra-strong steel pot with walls 6 inches think. It holds the fuel assemblies, the control rods and the water. Think of a spaghetti pot with the colander inside. After the zirconium tubes containing the fuel, the reactor pressure vessel is the primary means by which the radioactive material is contained. The RPV sits on a big concrete pedestal in the center of the giant concrete containment dome, which is substantially larger in diameter. The torus is connected to the containment dome. So even if the torus is breached, so long as the RPV is intact, the radioactive materials are contained (except what may be deliberately released in the steam when depressurizing the RPV).
Unfortunately, the fire may be a big deal. After nuclear fuel is used up in the reactor, the rods are removed and put in the spent fuel pool, which is a giant concrete swimming pool about 45 feet deep. (the news reports have erroneously called this a "reactor cooling pool". It's not). The fuel is usually covered by about 30 feet of water and the water in the pool is run through heat exchangers to keep it cool, since the fuel, while spent, continues to generate some heat. If the pool is not cooled (say, due to a total power failure) it will gradually heat up and could boil. The rate at which this occurs depends on the amount of fuel in the pool. I understand that the pool in Unit 4 had more spent fuel than the other units' pools. If the water should boil off to the point where the spent fuel is exposed to the air, it can catch fire. That would be a very bad thing, because it could result in the spread of contamination. As with the fuel melting in the reactor, the solution is to put water back in the pool, which is what they are trying to do now. I cannot tell from the current reports whether the fuel was on fire or it was something else.
Here is a Nuclear Regulatory Commission document with some decent diagrams, so you can see what I have been talking about.
http://www.nrc.gov/reading-rm/basic-ref/teachers/03.pdf