You will not be able to access the best product availble used by wildland fire agencies such as the NPS, USFS, BLM, CDF, etc. I believe that one is provided via a contract the BLM awards and administers. I used to get printouts from that one and the level of detail was beyond anything you can get on the Internet now. I understand that the maps are now merged with USGS topos, if necessary.
A coworker of mine and excellent fire investigator once located the exact ignition point of a 8,000 acre High Sierra wildfire that the public was claiming had to be human caused. Using ground pounding investigation techniques and the lightning locater map he was able to determine the time (9 days prior to the first report of the fire) and location of the strike that started the fire. The public interest was high as the fire nearly caused the evacuation of the town of Mammoth Lakes. It seem that lightning stuck a tree on the banks of the middle fork of the San Joaquin River and it burned for some time until it fell with the trunk on the east side of the river and the top on the west side. It then burned slowly from the middle until it reached each bank. It had dried considerabaly in the intervening 9 days and when the fire reached each bank we had two columns of black smoke (heavy fuel involved which indicates an intense fire), each on opposite sides of the river. We did not have the advantage of the river providing a natual barrier to the spread of the fire during initial attack. We lost the fire big time less than 24 hours later.
The lightning map showed two strikes in that area on that day. The investigator found the other strike and determined through burn patterns that it had burned a short time and then went out and was later burned over by the fire from the first strike. After the results of this investigation were made public many local "know at alls, who really don't know anything" pronounced the investigation useless.
When I worked on the Magdalena Ranger District of the Cibola National Forest back in the late 70's our Ranger Station was selected to have a lightning detector receiver during the first widespread experiement of the technology. All we saw was a bunch of red, yellow, and green lights flash on the side of the unit. It was hooked up to the phone lines and some computer in a big city interpreted the data from each receiver. It was the beginning of the technology now being employed nationwide.
By the way this is a good place to mention "holdover" or "sleeper" fires. The nine day period between the strike and the discovery of the fire in the case above is fairly common. I personally verified 18 days between a strike which I happened to see hit a tree on top of a ridge and my being dispatched to the fire it later grew into. I've met people who have had similar experiences of 30 days. I've chased fire starts in May where roots burned over winter with snowpack underneath the fire lines of moderate (100-300 acre) sized fires from the previous late September-early October period.