Your example, even if accurate (see other posts), doesn't show entanglement, as they could easily just have been created with those matching properties and went on their merry way.
For an example of action at a distance, consider a pair of photons, entangled with matching polarization (or not matching, but a known fixed relationship without having to be measured). And then apply the wizardry of a series of polarizing filters.
In this experiment, we have three polarizing filters. Starting with two filters, rotated by 90 degrees, they will stop any photon from passing, as a photon which passes through the first will be stopped by the second. Now, add a third filter between the first, rotated so that it is 45 degrees from each. A certain number of photons will now pass through the filter (the first weird part), showing that the center filter changed the orientation of photons which passed through the first filter. All photons which make it through to the other side will have their polarization matching that of the last filter. We haven't measured anything, but we know that the photon CANNOT have entered the filter setup with the polarization that it left, otherwise it would have been stopped by the first filter. As a statistical property, we are fairly comfortable with the idea that a polarizing filter, if it passes a photon, will in effect set the polarization of the photon to match the filter, and that statistically the closer to "aligned," the more likely it is to pass. Thus, some photons which passed the first filter make it through the second filter, and some that make it through the second filter likewise make it through the third filter, though if the second filter was not there none would pass. This isn't all that spooky.
NOW measure the polarization of the entangled photon, and what will you find? You will find that photons which passed through all filters have their entangled photons maintaining the relationship they had when they were created. This is true even though classic physics tells us that the photon that passed through the filter does not have the polarization that it started with (if it did, it would have been stopped by the first filter). This means that the entangled photon at some point changed its polarization, even though it passed through no filters. THAT is action at a distance, because changes made to the first photon after the pair were created have affected the second without apparent interaction. Also note that the effect is instantaneous, not bound by the speed of light.
Mike
Note - it must be pointed out that this DOES NOT enable instantaneous communication - you can't arbitrarily change the photon's orientation, only measure it to find out whether it is or is not oriented in a specific direction. The act of measuring will set 50% to that orientation and 50% to the opposite (well, 90 degrees off, technically), but you have no control over which 50%, and at the other end, their test will do the same. So someone looking at an entangled photon at the other end will be able to determine its polarization, but NOT whether that is due to the original being changed or simply having always been that way. Entanglement is "broken" at the first measurement, so any changes after that will not be reflected at the other end.