I just know that it's has something to do with the loudness levels being equal across the frequency spectrum, care to explain it further?
The X-axis is obviously frequency, and the Y-axis is power. Each curve on the graph is a "perceived loudness" level. The curves dip lower towards the middle-high end of the spectrum, because we hear those frequencies more easily. Each curve is labelled with a number equivalent to its perceived loudness level. If you look at each of those curves on the 1,000 Hz line (X-axis), you'll notice that they cross the power level that matches their perceived rating. For example, the 60 curve meets the intersection of 60 on the Y-axis (power), and the 100 curve meets the intersection of 100 on the Y-axis.
Things get compacted on either end of the spectrum, because more energy is required for us to perceive those frequencies as the same loudness level. Let's use that 60 curve as an example. We've already established that it crosses the power scale at 60 for 1,000 Hz. For us to perceive 30 Hz as the same volume as 1,000 Hz, it will need approximately 19dB more power (the 60 curve meets 30 Hz at approximately 79 on the Y-axis). 16 kHz needs about the same increase in power above 1,000 Hz to match in percevied volume (the 60 curve meets 16 kHz at approximately 80dB).
While there have been some flaws identified in the Fletcher-Munson curves (don't ask me what, because I don't know for certain), the principle applies to sound propagation/reproduction in any medium. Amplifiers and playback systems are designed using information similar to these (that's why we have the ".1" in 5.1, and things like "bass management" to make up for crappy mains in home theatre systems), and you could easily see similar results to the curve walking around outside with an SPL meter. Without that power increase, whether it come from electrical amplification or natural acoustic phenomenon, we won't hear things as being the same level.
Regarding flaws in the Fletcher-Munson curves: the original work establishing the curves was done in the 1920's and early 1930's under the supervision of Harvey Fletcher of Bell Laboratories. He was assisted by Fred Munson. At that time, making accurate measurements of the properties of sound was more difficult than it became later, the sources and instruments available were simply not as good as what we have now. Therefore, later researchers (Robinson and Dadson, for example) revised the curves, based on the improved measurements that they made.
The measurements are hard, not because of instruments (FM had sufficiently good instruments) but because the listeners have to tell you when two sounds at very different frequencies sound as loud as each other. And in the FM case, at sound levels up to 120dB (painful). If you try playing two sounds of very different frequencies figuring out when one is as loud as the other is tough. Fold in different people and their differing frequency responses. The scatter in the numbers is huge, and then you have to average them. Do you average the energies or the dB? etc. The three curves, the FM, the Robinson Dadson, and the ISO 2003 differ pretty significantly from each other. Which are "better"? Why do they differ so much from each other?(over 10dB in some places)? I do not think anyone knows, but it does indicate that one should take them with a certain grain of salt.