The present work does not obviate the need to follow-up NEAs with observations after they have been discovered. On the contrary, it is widely accepted that the current level of follow-up work is not yet at a satisfying level even for the northern hemisphere.
The problem is more important for smaller asteroids with absolute magnitude 20< H <22, corresponding to diameters for an average albedo. Most of these are discovered during close approaches, and end up being lost unless a chance rediscovery takes place in a later close approach. The strategy we have proposed is most applicable to these small bodies; for the larger objects with H<20 recovery should be considered mandatory if they can have impacts upon the Earth, even with small probabilities.
For even smaller objects, in the range of 22 < H < 25, not officially designated as Potentially Hazardous Asteroids, but still capable of causing a severe regional disaster, our technique of excluding the possibility of collisions without recovery might be the only option. Although the present method will usually allow us to deal with the hazard posed by objects down to the Tunguska size level (H=25), we may not always have another chance to observe associated Virtual Impactors before a threatening event. Indeed the discovery apparition may lead to a resonant return with collision without an opportunity to rule out the collision possibility prior to the supposed impact.
Another source of concern is the following. As the discovery rate continues to increase faster than the follow-up capability, especially at faint magnitudes, there is a real danger of creating Virtual Impactors at an unacceptably high rate, to the point that negative observation campaigns, their efficiency notwithstanding, would use too much observational resources.