Some relevant articles from the latest ASPJ:

Asymmetric Air Support

The land component is acutely aware that under current doctrine, it is apportioned/allocated CAS assets based only on CAS requirements.4 The word close in CAS does not imply a specific distance; rather, it is situational. The requirement for detailed integration due to proximity, fires, or movement is the determining factor, but this is becoming less and less relevant to what the ground component actually needs in order to serve as a stabilizing force. The need for CAS to deliver ordnance in close proximity to friendly forces is becoming a smaller factor in the current environments of Iraq and Afghanistan.

Over the last five years, fighter/bomber aircraft of the coalition air forces have evolved to become more than just strafing/bombing platforms. Granted, their targeting pods were designed to employ precision-guided munitions and reduce collateral damage, but the inherent capabilities of the pods have expanded their role into widely used and effective reconnaissance/surveillance. Unfortunately, the Air Force lacks the intelligence infrastructure to exploit the information garnered from the pods and other sources. The Air Force should have intelligence capability integral to the squadron, as did an RF-4 squadron, if it is going to fully exploit the intelligence gained from full-motion video (FMV) footage.5

In today’s operations, the land component has a great need for reconnaissance platforms; some people have even called it a “limitless hunger.”6 This need far exceeds the assets available to cover requirements, some of which are for armed reconnaissance to enable immediate strikes against the enemy during time-critical operations (e.g., indirect-fire setups and emplacement of improvised explosive devices). These requests may not involve close proximity to friendly forces or require detailed integration since no operations may be occurring at the proposed reconnaissance location. Even so, none of the current fighters in the Air Force’s inventory were designed as FMV reconnaissance platforms. The F-16C+ (Block 30), a reconnaissance-capable aircraft, replaced the RF-4 as the Air Force’s primary armed-reconnaissance platform, but its capabilities lack the real-time feed desired by the land component, which wants the real-time, FMV feed that it gets from aircraft equipped with the Remote Operations Video Enhanced Receiver (ROVER). Because the land component can’t fulfill reconnaissance-support requirements from organic assets or from surveillance and reconnaissance platforms, it now uses the JTAR process to request armed reconnaissance from traditional CAS assets. Although referred to as CAS to keep within doctrinal limitations, this is not CAS as the air component community would typically define it. Unfortunately, fighter units assigned to the two theaters of operations must provide CAS to the land component. This is where the friction starts.
Also a good (and related) piece on problems with ISR tasking:

I READ WITH GREAT interest Lt Col Michael Downs’s article “Rethinking the Combined Force Air Component Commander’s Intelligence, Surveillance, and Reconnaissance Approach to Counterinsurgency” (Fall 2008). I agree that the system needs reform but disagree with his proposal to use the close air support request process for intelligence, surveillance, and reconnaissance (ISR). Though faster than the air tasking order cycle, the joint tactical air strike request process used for close air support still does not allow the flexibility in execution required by the ground commander. Given that the Air Force has decided to apply theater ISR assets to tactical priorities, that commander must have the ability to shift those assets when priorities change. The combined air operations center (CAOC) must allow decentralized execution of ISR assets—particularly full-motion-video platforms—by delegating tactical control of platforms apportioned to Multi-National Force-Iraq (MNF-I) during execution. Doing so would speed the process of dynamic retasking and shorten the time required to respond to the ground commander’s shifting priorities and time-sensitive targets.